Measuring Plant Diversity: Lessons from the Field

  • 46 339 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview


This page intentionally left blank


Lessons from the Field

Thomas J. Stohlgren

Oxford University Press, Inc., publishes works that further Oxford University's objective of excellence in research, scholarship, and education. Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam Copyright © 2007 by Oxford University Press, Inc. Published by Oxford University Press, Inc. 198 Madison Avenue, New York, New York 10016 Oxford is a registered trademark of Oxford University Press The format and design of this publication may not be reproduced by any means without the prior permission of Oxford University Press. The text was produced by the National Institute of Invasive Species Science, Fort Collins Science Center, U.S. Geological Survey, and is in the public domain. Library of Congress Cataloging-in-Publication Data Stohlgren, Thomas J. Measuring plant diversity: lessons from the field / by Thomas J. Stohlgren. p. cm. 1. Plant diversity. 2. Botany—Methodology. 1. Title. ISBN 13: 978–0–19–517233–1 QK46.5.D58S86 2005 581.7—dc22 2005008060 98765432

To my lovely wife, Cindy, and to our charming children, Connor, Shannon, and Michael

This page intentionally left blank

Foreword There is no substitute for high-quality data in the scientific process. Hypotheses and models, as well as experiments, also play a central role in the advancement and refinement of scientific understanding. But neither have scientific value independent of quantified observations (data) on the behavior of natural systems. This is true in all fields of science, but particularly so for those disciplines in which the major phenomena cannot be experimentally manipulated, such as astronomy and cosmology (e.g., the formation and dynamics of stars, galaxies, and the universe as a whole), geology (the history and dynamics of the earth's solid structure), and ecology (the evolution and dynamics of the living organisms distributed over the earth). In spite of its central role in scientific progress, the collection of data on the properties and dynamics of ecological systems is often looked down upon as “soft science,” or as the early, primitive, “descriptive” phase of the science of ecology. The prestige of experimental science is epitomized by the emphasis on an experimental approach in research funding decisions. This is exemplified by the fanfare that accompanied the recent publication of ecological experiments heralded with the announcement from Peter Karieva that “ecological science has entered the big leagues, the world of particle accelerators and orbiting telescopes.” The status and credibility accorded such experimental results have led some ecologists to question or dismiss field observations that were inconsistent with the new experiments. Two types of field observations have

viii recently been called into question. First, the widely observed pattern of plant diversity being lower in areas with high plant productivity, often associated with fertile soils or the addition of fertilizer, than in areas with intermediate productivity was called into question by experimental results that seemed to show that plant productivity increased monotonically as the number of plant species increased. Various efforts have attempted to explain or dismiss the “anomalous” patterns found in the field data. The second type of field data criticized for being inconsistent with experimental results was the observations (including some on which this book is based) that areas with high numbers of native species also have high numbers of exotic species. These observations conflicted with new experimental results that seemed to show that areas with high numbers of native species were more resistant to invasion by exotic species, and consequently had lower numbers of exotic species than areas with fewer native species. The field data not only failed to match the experimental results, but were also inconsistent with an old and widely held belief (hypothesis) that high diversity conferred resistance to invasion, as we learned from Charles Elton in 1958. The fact that this familiar hypothesis was supported by the results of several new experiments called the validity of the field observations into question and led to efforts to explain how the field studies could produce such misleading results. While on first inspection the new experimental ecology on the model of “particle accelerators and orbiting telescopes” seemed to invalidate the ecology of quantitative data based on field measurements, this initial conclusion has been reversed. My extensive reevaluation and analysis of these experiments has led to the discovery of numerous problems with the experimental design, analysis, and interpretation of both the diversity-productivity experiments and the diversity-invasibility experiments. In two such examples, the inconsistency between the field data and the experiments reveals the flaws in the experiments, not vice versa. In the final analysis, it is quantitative data based on observations of “real-world” systems that represent the truth to which both hypotheses and experimental results must be compared. The limitations imposed by experimental controls produce results that are typically relevant to only a small portion of the full range of conditions (e.g., the mean, range, and spatiotemporal variability of multiple environmental factors) experienced by natural ecological communities. This book addresses the collection and analysis of field measurements of plant communities and includes data that are relevant to and that have played a role in the resolution of the two controversies discussed above. The ecological patterns analyzed and discussed here illustrate the power of high-quality ecological data to the advancement of scientific understanding. While some recent ecological experiments have failed to live up to the high standards of particle accelerators, the methods and data discussed in this book illustrate the commonality between quantitative field ecology and the hightech science based on orbiting space telescopes. Two parallels merit

ix brief elaboration. Just as deep-field telescopes allow the detection and analysis of individual stars and galaxies of different ages, different stages of cosmogenesis, and differing initial or current interstellar conditions, so the selection of field sampling locations allows analysis and comparison of plant communities at different stages of development, with different initial and current environmental conditions. A second similarity relates to the collection of cosmological data at multiple frequencies along the electromagnetic spectrum, from X-rays to visible light to infrared radiation. Just as these multiple frequencies allow sampling of different types of cosmogenic processes across a range of spatial resolutions, so the multiscale sampling methods described in this book allow the detection of biological patterns caused by ecological and evolutionary processes that occur at different spatial (as well as temporal) scales. It is critical that quantitative field ecology be based on the recognition that different types of processes (e.g., competition versus dispersal) operate at different spatial and temporal scales. The inevitable consequence of failure to make this distinction is that the effects of a given process can bé undetectable in samples collected at an inappropriate scale. The danger of ignoring the process-scale relationship is that processes that remain undetected by an inappropriately scaled sampling design are often assumed to be unimportant for regulating ecological phenomena. Such errors misdirect the scientific process and may lead to serious errors in resource management and conservation. Tom Stohlgren has done an excellent job in raising these issues in the context of quantitative field ecology, building on the strong tradition of Frederick Clements, Robert Whittaker, John Curtis, and other great plant ecologists of the 20th century. The sampling methods that are developed, evaluated, and applied in the pages of this book provide a sound foundation for a quantitative field ecology that can advance our understanding of fundamental ecological and evolutionary processes. They also provide a standard for testing hypotheses and for evaluating the relevance and validity of hypotheses and experimental results. This book provides a good introduction to field sampling for beginning plant ecologists, as well as a fascinating set of results and analyses that will stimulate and challenge experienced ecologists. —Michael Huston San Marcos, Texas

This page intentionally left blank

Preface Most textbooks on measuring terrestrial vegetation have focused on the characteristics of biomass and cover, and on the density or frequency of dominant life forms (trees, shrubs, grasses, and forbs), or on classifying, differentiating, or evaluating and monitoring dominant plant communities based on a few common species. Sampling designs for measuring species richness and diversity, patterns of plant diversity, species-environment relationships, and species distributions have received less attention. There are compelling, urgent reasons for plant ecologists to do a far better job measuring plant diversity in this new century. Rapidly invading plant species from other countries are affecting rangeland condition and wildlife habitat, placing more plant species on threatened and endangered species lists and increasing wildfire fuel loads. Attention has shifted from the classification of plant communities to accurately mapping rare plant assemblages and species of management concern to afford them better protection. More ecologists, wildlife biologists, and local and regional planners recognize the value in understanding patterns, dynamics, and interactions of rare and common plant species and habitats to better manage grazing, fire, invasive plant species, forest practices, and restoration activities. Thus revised and new sampling approaches, designs, and field techniques for measuring plant diversity are needed to assess critical emerging issues facing land managers. This book offers alternatives to the approaches, designs, and techniques of the past that were chiefly designed for dominant species and other purposes. I focus on field techniques that move beyond classifying, mapping,

xii and measuring plant diversity for relatively homogeneous communities. This book complements methods for measuring the biomass and cover of dominant plant species. Most species are sparse, rare, and patchily distributed. It empowers the reader to take an experimental approach in the science of plant diversity to better understand the distributions of common and rare species, native and nonnative species, and long-lived and short-lived species. There are six parts. Part I introduces the problem: plant diversity studies are difficult to design and conduct, in part because of the history and baggage associated with the evolution of plant ecology into a quantitative science. Issues of scale, resolution, and extent must be effectively commandeered. Part II implores the practitioner to take an experimental approach to sampling plant diversity with a clear understanding of the advantages and disadvantages of single-scale and multiscale techniques. Two case studies (actual field investigations) demonstrate how to test and assess various field techniques. Part III focuses on scaling plant diversity measurements from plots (or local field observations) to landscapes. Here, five detailed case studies are used to arm the practitioner with model applied studies of plant diversity from issue formulation, through methods and statistical analysis, to cautiously interpreting results. The case study methods employ the select multiscale plot design with advantages and disadvantages, but alternative multiscale methods could easily have been employed. Regardless of the specific study methods selected, the general approaches might be used and improved by others to measure plant diversity to meet various management needs. Part IV provides a brief introduction to modeling plant diversity in relation to environmental factors. Examples of common nonspatial (correlative) and spatial analyses are explained. Part V introduces the concept of measuring temporal changes in plant diversity at landscape scales followed by a case study designed to collect the necessary baseline data to monitor plant diversity. Part VI discusses research needs to better understand changes in plant diversity in space and time. The intended audience for the book includes upper division college students, graduate students, field ecologists, resource managers, landowners, range conservationists, and others. I specifically target observational studies of plant diversity and patterns of plant diversity at landscape scales, the survey and monitoring of forest understory and grassland plant species, species-environment relationships and species-area relationships, the role of rare habitats in maintaining plant diversity patterns, typical statistical analysis techniques, and modeling spatial patterns of plant diversity. This book is not a complete text on comparative field methods or biometry. It doesn't cover controlled vegetation experiments or theoretical ecology. It is particularly weak on aquatic vegetation techniques and nonvascular plant sampling techniques, though many of the proposed techniques might be easily adapted for those uses. The book is limited on the many theories

xiii and causes of plant diversity, since these topics go well beyond observational studies. Measuring Plant Diversity is a logical extension and synthesis of a body of work that spans about 10 years, including more

than two dozen peer-reviewed journal articles, book chapters, and lecture notes that I have written, with the help of many colleagues. I cite and acknowledge my more than 60 coauthors who have contributed to jointly published papers over the years. My coauthors, in alphabetical order, include Craig Allen, Richard Bachand, Bill Baker, Dave Barnett, Jill Baron, Michael Bashkin, Jayne Belnap, Joe Berry, Dan Binkley, David Buckley, R. Busing, Tom Chase, Geneva Chong, Scott Collins, Michael Coughenour, Nolan Doesken, Charles Drost, Paul Evangelista, Maurya Falkner, Leandro Ferriera, Curt Flather, Paula Fornwalt, Jim Grace, Sue Grace, Laurie Huckaby, Mohammed Kalkhan, John Kartesz, Merrill Kaufmann, Margot Kaye, Kate Kendall, Tim Kittel, Michelle Lee, Jesse Logan, John Moeny, Dennis McCrumb, Greg Newman, Yasuhiro Onami, Paul Opler, April Owen, Bill Parton, Bruce Peterjohn, Betsey Pfister, Roger A. Pielke, Jr., Roger A. Pielke, Sr., Kelly Redmond, Robin Reich, Kelly (Bull) Rimar, J. Rodgers, Mike Ryan, Yuka Otsuki, James F. Quinn, Michael Ruggiero, Lisa Schell, Sara Simonson, Yohon Son, Jason Stoker, Ken Stolte, Kuni Suzuki, Hidiru Suzuki, Brian Vanden Houvel, Tom Veblen, Cynthia Villa, and Gary Wagonner. I have received their full support and encouragement throughout this process. I am the storyteller, but it is our story, and I will never be able to thank them enough for their guidance and training. I tell the story of Measuring Plant Diversity by relying heavily on examples. The “lessons from the field” aspect comes from hiking and observing, testing and evaluating, and trial and error. I provide general techniques and methods that can be used directly or easily modified by most students to meet their needs, and case studies that have both general and specific applications. I provide copious tables and figures adapted from previous publications because real field data are important in the learning process. I supply many stand-alone case studies, and “lessons” in the form of photographs and figures to pass on a few hints, raise questions and issues, and stimulate discussion and thought. I believe that “example is the best teacher” —and that the “teacher isn't always right.” Wise students will use the lessons in this book as a starting place to address a few questions rather than as a place to look for all the answers to some final exam. My specific objectives are to (1) provide a basic understanding of the history of design considerations in past and modern vegetation field studies; (2) demonstrate with real-life case studies the use of single-scale and multiscale sampling methods, and statistical and spatial analysis techniques that may be particularly helpful in measuring plant diversity at landscape scales; and (3) address several sampling questions typically asked by graduate students and budding field ecologists. My ultimate goal is to encourage additional studies of plant diversity—arming the next generation of ecologists with a better

xiv understanding of the history of the development of vegetation science as it relates to plant diversity, a rationale for taking more of an experimental approach in designing large-scale plant diversity studies (we're all very new at this), and the early beginnings of a better toolkit for future advances in the field of measuring plant diversity. I welcome all feedback, comments, and suggestions.

Acknowledgments In addition to my valued coauthors, equally valuable field and laboratory assistance over the past 10 years has been provide by Krista Alper, Marcell Astle, Steve Bousquin, Debbie Casdorph, Jean Marie Ederer, Rick Edwards, Helen Fields, Emily Galbraith, Randy Griffis, Michele Hart, Kate Healy, Catherine Jarnevich, Amy Johnson, Susan Klimas, Tom and Jeane Leatherman Alicia Lizarraga, Max Medley, John Moeny, Stephanie Neeley, Lisa Nelson, Anne Overlin, Nate Pierce, Amy Randell, Dan Reuss, James Self, Rick Shory, Elizabeth Smith, Sean Stewart, Connor Stohlgren, Laura Stretch, and Alycia Waters. In addition, the book's final references, tables, and figures received help from Nate Alley, Dave Barnett, Geneva Chong, Paul Evangelista, Mohammed Kalkhan, Deb Guenther, Greg Newman, Sara Simonson, and Alycia Crall— and I can't thank them enough. Advice from Indy Burke, Norita Chaney, Phil Chapman, James Detling, Bill Gregg, Alan Hastings, Bruce Van Haveren, Michael Huston, Linda Joyce, John Kartesz, Bill Laurenroth, Mark Miller, Dennis Ojima, Michael Palmer, and Marcel Rejmánek has been greatly appreciated. Logistical support was provided by the U.S. Department of the Interior, National Park Service, National Biological Survey, National Biological Service, Bureau of Land Management, U.S. Geological Survey (the Midcontinent Ecological Science Center, now the Fort Collins Science Center), and Natural Resource Ecology Laboratory at Colorado State University. To all, I am eternally grateful.

This page intentionally left blank

Contents Foreword


PART I: THE PAST AND PRESENT 1. Introduction 2. History and Background, Baggage and Direction 3. A Framework for the Design of Plant Diversity Studies

3 15 46


Single-Scale Sampling Multiscale Sampling Comparing Multiscale Sampling Designs: Taking an Experimental Approach Case Study: Comparing Rangeland Vegetation Sampling Techniques

73 92 111 118

PART III: SCALING TO LANDSCAPES 8. Case Study on Multiphase and Multiscale Sampling


xviii 9. Case Study: Designing a Monitoring Program for Assessing Patterns of Plant Diversity in Forests Nationwide 10. Case Study: Patterns of Plant Invasions in Forests and Grasslands 11. Case Study: Evaluating the Effects of Grazing and Soil Characteristics on Plant Diversity 12. Case Study: Assessments of Plant Diversity in Arid Landscapes

159 171 191 218

PART IV: MODELING PATTERNS OF PLANT DIVERSITY 13. Nonspatial Statistical Modeling of Plant Diversity 14. Spatial Analysis and Modeling

239 254

PART V: MONITORING PLANT DIVERSITY 15. Concepts for Assessing Temporal Changes in Plant Diversity 16. Case Study: Monitoring Shifts in Plant Diversity in Response to Climate Change

273 287

PART VI: RESEARCH NEEDS 17. Case Study: Testing a Nested-Intensity Sampling Design 18. Quantifying Trends in Space and Time Glossary References Index

307 323 341 343 375


This page intentionally left blank

1 Introduction Despite several solid attempts to compile standard methods for vegetation sampling (Bonham 1989; Daubenmire 1968; Elzinga et al. 1998; Mueller-Dombois and Ellenberg 1974), many sampling techniques dealing specifically with plant diversity have not been widely accepted or generally applied. Any review of methods sections in plant ecology journals will reveal a myriad of approaches for selecting sampling sites, plot shapes and sizes, and sample sizes, and many different patterns of sampling. Few studies have collected directly comparable data on plant species richness or diversity because of the many subjective decisions made during the course of each study, from the sampling design phase, through field-testing techniques, through statistical analyses, modeling, and design modifications to application. There are two primary reasons why standard methods for plant diversity sampling have not been as widely accepted in the same way that standard methods for water sampling and analysis or for soils analysis have been adopted. First, specific study objectives may vary greatly among investigators of plant diversity. Some studies are designed to address site-specific management or research needs, such as vegetation mapping, wildlife habitat analysis, or specific grazing or postfire effects, etc. Specific sampling designs and field techniques might be required for each objective. Second, as noted by Hinds (1984, p. 11), “it is a world of unique places”—it has been difficult for plant ecologists to design offthe-shelf vegetation sampling protocols or monitoring techniques for general use. Vegetation types vary from desert scrub to alpine tundra, from tropical rainforests to boreal forests, and from aquaticlily



pads to giant sequoias. Understory plant diversity can vary from sparse and patchy under heavily canopied forests to rich and productive in wet meadows and riparian zones. Vegetation structure at each site may be affected by the topography, geology, soils, hydrology, land use and disturbance history, and the history of dispersal, establishment, growth, and survivorship of the cornucopia of species at each site. It is no wonder that generalized techniques might fail for specific purposes and unique sites, or that specific techniques might fail for various vegetation types and general applications. This seemingly chaotic status of vegetation science should not prevent plant ecologists from improving the level of standardization among field sampling techniques for evaluating plant diversity. Understanding the nature of the problem is an important first step in designing solutions. I begin with a brief evaluation of the nature of the problem.

Why Conducting Plant Diversity Studies Is Difcult There are several well-known difficulties in conducting plant diversity studies, including taxonomic difficulties, phenology difficulties, and problems of species rarity. These difficulties are often deterrents to graduate students and young ecologists, and they shouldn't be. They should be viewed as challenges with many potential solutions.

Taxonomy Difculties There are recognized problems in identifying plant species in the field. There are large numbers of species in the species pools of most landscapes. For example, Rocky Mountain National Park in Colorado (108,000 ha) contains more than 1000 plant species, and more than 100 plant species from other countries (i.e., nonnative or exotic plant species). The State of Hawaii contains about 1900 species of angiosperms (flowering plants), with half of the species coming from other countries ( The Jepson manual of higher plants in California recognizes 5862 native and 1023 nonnative naturalized species from other countries (Hickman 1993). Some species (many examples), genera (e.g., Carex), and families (e.g., Poaceae) are particularly difficult to identify in the field because of small or variable plant parts. Many species contain subspecies and varieties. This explains the wide range of plant characteristics that investigators are expected to differentiate. A species named “polymorphous” is particularly discomforting to field Hint: It is important to record the finest taxonomic resolution possible in the field—you can always lump species, but you can't split after the fact.



ecologists. Phenotypic variation is common for broadly distributed species. For example, the silver sword in Hawaii varies so much in form from island to island and habitat to habitat that only the best field naturalists can differentiate them. The Hawaiian silver sword alliance contains about 30 species in three genera (Argyroxiphium, Dubautia, and Wilkesia) with a phenomenal range of anatomical, morphological, and ecological adaptations, but are very closely related based on biosystematics and molecular studies ( Another general problem is that taxonomic keys are far from perfect. Many taxonomies are outdated. Local floras often differ from regional and national floras (e.g., the Flora of North America Project). There is often confusion and disagreement over two floras in the same state [e.g., Munz and Keck (1959) and Hickman (1993)]. Additional taxonomic problems occur when species, subspecies, or varieties are lumped or split, or when they are renamed as previously named taxa or new taxa. New names can be a daily occurrence. A newly recognized problem is that many floras are incomplete due to the rapid invasion of nonnative species into nearly every landscape. Discriminating between native and exotic congeners can be very difficult. The major problem with taxonomy, however, is a lack of investigator expertise in the field (Figure 1.1). Many plant ecologists receive little formalized training in plant systematics, relying on one or two undergraduate classes in taxonomy. Typically these classes involve a survey of plant family characteristics and the identification of 50 to 100 species in full bloom and with all the distinguishing characteristics. Thus many young ecologists may be unprepared for studies in species-rich environments or for landscapes and regions with large and complex species pools. There are many possible solutions to these taxonomic issues. The most obvious is to better train plant ecologists in basic field taxonomy. There are ongoing efforts to computerize taxonomic keys, including “polyclaves,” which can identify plant species from a few field characteristics. Polyclaves are already available for the plant families of the world ( or, and many are being created for state floras. Note that computerized taxonomic keys often have many of the imperfections of written manuals. Still, there is hope that future computerized keys will be more quickly updated and will provide rapid searches for “synonyms” (all historically used names for a given species) and remote computer assistance via satellites, including photographs, line drawings, and online taxonomic assistance.

Phenology Difculties Once in the field, there are several phenological difficulties that must be overcome. Surveys and monitoring are expensive, with the most expensive aspect being the cost of getting field crews to and from the sampling sites. In large landscapes, investigators can rarely afford to visit the same



Figure 1.1. Lesson 1. Plant diversity studies require well-trained, well rewarded taxonomists. The greatest sampling designs, field methóds, and statistics are worthless without botanists, taxonomists, and natural history experts. Always indicate the “authority” you are using when collecting so that the meaning and synonymies of the name can be reconstructed. Befriend taxonomists! sampling site twice. A common concession made in designing plant diversity field studies is to survey sites while most plants are, or have recently finished, flowering (often termed “peak phenology”). Of course, plant species do not perfectly synchronize flowering at any given site, so the plan is generally to capture as many plant species with flower parts as possible. Extensive botanical experience and observations of seasonal climate are helpful, but a naturalist's eye and extensive hiking are adequate substitute skills in many cases. In some ecosystems, multiple blooms are common. The shortgrass steppe in Colorado and many arid ecosystems contain some species that flower in early spring and some that flower in the late summer or fall, associated with periods of intense precipitation. In such cases, sampling sites have to be visited at least twice a year. In areas protected from intensive grazing, fire, and other disturbances, sampling in the fall often allows for identification of most early (dried flower parts) and late flowering species. The last caveat deserves more attention. Plant parts in nearly all areas are seriously compromised by herbivory from domestic livestock and wildlife. The role of insects and pathogens that attack plant structures also cannot be



Hint: The solution to finding many partially eaten plants is to search outside plots for more complete plant specimens. Plants play a great game of “hide and seek” from herbivores, and it is usually possible to find more complete plant specimens nearby. Where all the specimens of a species are immature, returning to the site at a later date is important. It is wise to collect and press partial specimens in a working herbarium to compare later to complete specimens. underestimated. It is not uncommon to have the majority of plants grazed or browsed in a study area.

Difculties with Species Rarity Another well-documented problem with plant diversity studies is the uneven distribution of the number of individuals in each species in an area, which complicates field measurements and sampling designs. Species abundance curves are almost always inverse J (or negative exponential) curves, with a few species having many individuals and the vast majority of species represented by very few individuals. In several field studies in the Central Grasslands, Rocky Mountains, and Colorado Plateau, we have documented that about 50% of the species encountered in 0.1 ha plots have less than 1% foliar cover (Stohlgren et al. 1997b, 1998d, 2000b). Most species are locally sparse. Like species abundance curves, species frequency curves tend to have an inverse J shape (Figure 1.2). Of the 550 understory plant species recorded in the 309 plots (0.1 ha each) distributed in the 850,000 ha Grand Staircase-Escalante National Monument in Utah, 189 appeared in only 1 or 2 of the plots, while only 36 species were found in 50 or more plots.

Figure 1.2. Data from 309 0.1-ha plots in Grand Staircase-Escalante National Monument in Utah (unpublished data). Species codes (PLANTS database; are shown for selected species.



A consequence of, or correlate to, rarity in species abundance curves and species frequency curves is that most plant species have very patchy distributions. In another example, in a random selection of four 0.1 ha plots in four prairie types in the Central Grasslands, only one of nine species, on average, was found in all four plots in each vegetation type (Stohlgren et al. 1998c). Many species in a vegetation type occurred in only one of the plots. In 1 m2 subplots in the same prairie types, a few species were commonly found together, but few of the many locally rare species could be consistently found in the same subplots. Patchy distributions and natural spatial variation are compounded by temporal variation (Huston 1999; Huston and McBride 2002). Temporal variation can be caused by immigration, emigration, seed germination, extirpation, speciation, and extinction. Immigration is very common, especially considering invasive exotic species. Plant species can germinate soon after field crews leave a site, with seed banks providing sudden species enrichment. Conversely, established species can desiccate to unrecognizable forms just prior to a field crews visit (i.e., local extirpation). Emigration, speciation, and extinction are generally much slower processes that are beyond the scope of most plant diversity studies, but they may be important in some areas and should not be discounted outright. The solutions to problems of local rarity are to search larger areas, search well, and search often. I will discuss later ways to evaluate how large of an area needs to be searched and how often one should revisit sampling sites. Evaluating how well a site is searched is also important.

Why Designing Plant Diversity Studies Is Difcult The difficulties just described in conducting plant diversity studies make it difficult to design such studies. Designing plant diversity studies for large landscapes is difficult for many reasons. The difficulties result from high spatial and temporal heterogeneity, which occurs at multiple spatial scales—from the rarest and tiniest of seeds in the seed pool to global patterns of diversity—with many species of plants, animals, and diseases rapidly dispersing Hint: The most important question a plant ecologist can ask is, “What did we miss by sampling?” Ecologists often report “what they've captured” (e.g., the number of species in a plot or group of plots) without evaluating the completeness of the work. There are many ways to evaluate completeness by extending the size of plots or using multiscale plots, or by adding time to searches for species. These topics are described later in the text (Figure 1.3). Lesson 2— Always ask, “What did we miss?”



Figure 1.3. Photograph by Paul H. Evangelista. Used with permission. throughout the biosphere. Imagine designing plant diversity studies in light of the complexities described below.

Plant-to-Plant Neighborhood-Scale Variability At the neighborhood scale, two plant species can be (1) direct competitors, competing for the same resources; (2) dependent species, such as mosses that grow on particular tree species; or (3) complementary species, which use resources at different times or in different spaces (Huston 1999; Walter 1964, 1971). Furthermore, these dependencies may overlap at various growth stages, or in various habitats (Mueller-Dombois and Ellenberg 1974). When even the smallest plant dies, some resources may be released for a replacement by the same species, another species, or neither. Plants die for many reasons: age, disease, competition, and herbivory are typical causes. The cause of death may be drastic or slow. Life spans vary for plant species living in the same vicinity. Annual grass species can share a site with perennial herbs, moderately long-lived shrubs, and very long-lived trees. Local plant replacements can be sporadic over time, resulting in complex patterns of diversity at many spatial scales. Large numbers of individuals at a site may reduce the chance of local extirpation of a plant species, but it is often difficult to get accurate estimates of the number of spindly annual grasses or diffuse bunch grasses with dead or partially dead centers. It is equally difficult to identify individuals of sod-forming grasses or resprouting shrub and tree species. Defining “individuals” sometimes bothers plant ecologists. The abundance, diversity, variation, and viability of seed banks, and the timing of germination can also complicate the study of local plant species diversity.



Plot- to Landscape-Scale Variability We can envision a landscape filled with thousands to millions of asynchronous plant neighborhoods superimposed on complex topographic variation (slopes, aspects, elevation), physical factors (soil texture, geology), and environmental gradients (moisture, soil nutrients, solar radiation). Small-scale, frequent disturbances (e.g., tree falls, small mammal excavations) can be superimposed on larger-scale, less-frequent perturbations (e.g., volcanism, fire, insect outbreaks, rapid climate change; see Figure 1.4) that vary spatially and temporally (Huston 1999; Huston and McBride 2002; Pickett and White 1985; Stohlgren 1994). Spatial variation in plant diversity can be enormous: plant species lists in 0.1 ha plots just 100 m apart in grasslands in the Rocky Mountains overlap 30–80% (60% on average; Stohlgren et al. 1999b). Most species occur in multiple vegetation types (Figure 1.5). Landscapes represent a mix of species that have migrated to the site over thousands of years under various climate and disturbance regimes. For example, one 0.1 ha area in the middle elevations of Rocky Mountain National Park, Colorado, might include relict limber pine (Pinus flexilis) from the last

Figure 1.4. Lesson 3. Previous vegetation patterns influence seed banks and fire, which in turn influences the complex patterns of plant diversity at multiple scales in the post-burn stand, landscape legacies help to create complex spatiotemporal patterns of plant diversity in most natural ecosystems. Photo reproduced with permission of Geneva Chong.



Figure 1.5. Data from 24 0.1-ha plots in five vegetation types in Rocky Mountain National Park, Colorado. Most plant species were found in more than one vegetation type. A more detailed investigation would find very few species strictly limited to one forest type. ice age (10,000–12,000 years ago), ponderosa pine (Pinus ponderosa) that arrived approximately 6000 years ago, species of cacti that may have migrated to the site during a warming period (3000–5,000 years ago), various native plant species that arrived intermittently or in groups over the past 12,000 years, and up to 100 nonnative plant species that arrived from other countries in the past 150 years. In the same area of the park, there is overlap in species composition (typically 20–40%) among vegetation types and complex ecotones between vegetation types, making it impossible to delineate exact boundaries of vegetation types. Spatial and temporal variation in plant turnover, seed banks, seed dispersal from very distant sites (by wind, birds, currents, etc.), and the timing of germination add to the complexity in patterns of plant diversity in most landscapes. A vegetation survey is sometimes thought of as a snapshot in geologic time. It might be more realistic to think of a vegetation survey as a collage of species from multiple geologic times experiencing potentially rapid changes. The changes can occur in large areas from flooding, fire, or insect outbreaks. Alternatively, the changes can occur in small areas from small mammal excavations, tree fall, or any small-scale disturbance. It is obvious that conventional sampling strategies that randomly select a few small study plots in relatively rare homogeneous study units may not be adequate to describe plant diversity at landscape scales (Hinds 1984; Stohlgren 1994).

Landscape- to Regional-Scale Variability We also can envision a region filled with hundreds to thousands of asynchronous landscapes superimposed on complex topographic variations (longitude, latitude, elevation), climate factors (precipitation, temperature), geology, and longer environmental gradients (moisture, soil nutrients, solar radiation) than seen at landscape scales. The physical environment and



living organisms are rarely uniformly or randomly distributed. Instead, at all scales, resources are aggregated in patches, or they form gradients or other kinds of spatial structures (Legendre and Fortin 1989). Measuring plant diversity at landscape to regional scales is further complicated by extremely long lag times in evolution, migration, and speciation. Ancient riverbanks may still contain Equisetum (horsetails), a primitive vascular plant that evolved hundreds of millions of years ago, sharing the habitat with newly arrived modern angiosperms from other countries [e.g., the invasive purple loosestrife (Lythrum salicaria)]. Migration patterns have been vastly accelerated by modern transportation such as freightliners, airplanes, railroads, and trucks. Many geographic barriers to migration and speciation are easily breeched by the horticultural industry. Reichard and White (2001) report that between 57% and 65% of the flora of Australia was intentionally introduced for horticultural reasons. Meanwhile, modern land use, primarily agriculture and urbanization, often leaves small islands of natural vegetation in a sea of transformed land. The small refuges may support smaller, more isolated populations of plants and animals, setting the stage for additional phenotypic variation, speciation, or extirpation. Plant ecologists have long realized that plant populations and species are clumped (aggregated) in distribution rather than randomly located on the landscape (Ashby 1948; Greig-Smith 1964). Recognizing the patchy distribution of plant species from the neighborhood scale to the regional scale is the first step in addressing sampling design considerations.

Long-Term Changes in Plant Diversity Climate and weather history and patterns provide an analogy for understanding long-term changes in plant diversity. Dominant plant species in a region may reflect long-term seasonal patterns in temperature and precipitation. Other species in the region may have been established during various cycles (e.g., El Niño years, severe droughts) and lingered through suboptimal climate years. Still other species may have been established after extreme events such as floods or in high fire years as a result of extreme drought and high winds. Thus understanding long-term mean climatic conditions alone may not necessarily aid in the understanding of long-term changes in plant diversity. Many other factors, such as dispersal, migration, herbivory, disturbance, edaphic characteristics, competition, and seed bank viability, among others, affect long-term changes in plant diversity. Subtle changes in plant diversity at landscape scales cannot be detected without a long-term record (Strayer et al. 1986). Because long-term studies are rare, ecologists have only a cursory understanding of coupled spatial and temporal variability (Haury et al. 1978; Levin 1992; Powell 1989; Steele 1989). Thus, far more long-term studies, innovative sampling and monitoring strategies, and new analytical and modeling tools may be needed to evaluate long-term changes in plant



Figure 1.6. Lesson 4. Plant diversity (northwestern Wyoming) is patchy at multiple spatial scales in all natural landscapes–get used to it!

Figure 1.7. Lesson 5. Tamarisk (Tamarix spp., salt cedar) invading an arid landscape in southern Utah. Plant invaders like salt cedar can radically alter vegetation structure, water and nutrient cycles, soil chemistry, and local plant diversity. More invaders are on the way! (Photograph by Paul H. Evangelista, used with permission.)



diversity (Stohlgren 1994). Plant species are rapidly invading from other countries (Dukes 2002; Floerl et al. 2004; Gilbert and Lechowicz 2005; Mack et al. 2000; Stohlgren et al. 2005a,b,c). The obvious solution is to become extremely clever about evaluating plant diversity at multiple spatial scales and over time.

2 History and Background, Baggage and Direction In this chapter I provide a very brief history of plant ecology to focus on how previous ecologists have influenced the ways we typically measure plant diversity today. I draw particular attention to the authors of two text-books, Rexford Daubenmire, and Dieter Mueller-Dombois and Heinz Ellenberg, because they seem to best reflect the development of many current plant diversity field methods (Barbour et al. 1999). Lastly, I discuss the general direction provided by past plant ecologists, and on the “baggage” of older ideas—how inertia developed and persists in modern plant ecology regarding measuring plant diversity.

History and Background In the Beginning To understand modern approaches to the measurement of vegetation, it is important to provide a brief history of plant ecology. Many writers have provided more detailed histories of the people, places, culture, and schools of thought emanating from our scientific ancestors (e.g., Barbour et al. 1987; McIntosh 1985; Tobey 1981). I focus instead on the influence early naturalists and plant ecologists had on paradigms and sampling designs. One concept binds ecologists: the quest to understand species-environment relationships. Earlier peoples recognized species-environment relationships


History and Background, Baggage and Direction

in everyday life. McIntosh (1985) noted that Theophrastus, from ancient Greece, described mangrove forests related to saltwater habitat. Charles Darwin's The Origin of Species (1859) is the defining volume of observations, theory, conjecture, and wonder about species-environment and evolutionary relationships. Evaluating the primary effects of Darwin and his predecessors on sampling designs is difficult. In carefully documenting both broad-scale patterns of vegetation and wildlife in several countries, and the inseparable links between biotic and abiotic causes and effects, Darwin saddled future ecologists with the prospects of complex, multivariate explanations for the patterns and processes of plant ecology. The major force in early plant ecology was Friedrich Heinrich Alexander von Humboldt (1769–1859) from Prussia (Figure 2.1). Humboldt and others collected 60,000 plant specimens from throughout Central and South America from 1799 to 1804, later producing 14 volumes on the botany, physiognomy, and “associations” (vegetation types) of the region (Barbour et al. 1987). He described many associations with respect to interrelationships among latitude, elevation, and temperature. He believed that all things were connected by linkages of causes and effects. He published a vegetation classification system in 1806 based entirely on growth form, ignoring taxonomy for the most part. Humboldt established two important paradigms in plant ecology: (1) that vegetation can be described in “associations” based primarily on dominant species, and (2) that species and associations can be understood in terms of measurements of relatively few topographic and environmental factors. How did Humboldt influence vegetation sampling procedures? Most naturalists and plant geographers in the early 1800s continued

Figure 2.1. Schematic diagram of selected historical figures who influenced modern plant diversity research.

History and Background, Baggage and Direction


to use “searching” techniques—collecting and cataloging plant specimens, and taking careful notes of key environmental factors. Humboldt drew attention to the need for careful, systematic observations (“biological arithmetic”) to show, for example, the vertical zonation of vegetation. Ecological plant geography was born (McIntosh 1985) as Humboldt recognized the need to accurately measure and sketch natural phenomena. Influenced by a subset of Humboldt's ideas on plant associations, J. F. Schouw (1789–1852) emphasized temperature as the single most important factor in determining plant distributions. He described vegetation associations by attaching the suffix “etum” (meaning “community of ”) to genus names of the dominant species. The suffix is sometimes used today. Although many taxa may have been recorded at each site, attention may have been inadvertently drawn away from the understory and locally rare species with an overemphasis on the few dominant species (and their relationship to temperature) at each site. Anton Kerner von Marilaun (1831–1898) described plant life of the Danube Basin with detailed descriptions of vegetation associations and orderly succession. Kerner von Marilaun emphasized order: “The horizontal and vertical structuring of large plant communities is by no means accidental…Every plant has its place, its time, its function, and its meaning. In every zone, plants are distinct groups which appear as either developing or as finished communities, but never transgress the orderly and correct composition of their kind” [Conard (1951) as quoted in McIntosh (1985)]. Kerner was typical of most 19th-century plant geographers and the prevailing beliefs of order in nature and a world filled with distinct plant associations. Plant geography influenced many future plant ecologists in the design of studies, the development of quantitative field methods, and the interpretation of results. Danish botanist Johannes Eugenius Warming (1841–1924) recognized the linkage of physiology and morphology. He drew attention to the important role of soil moisture in vegetation patterns. He summarized vegetation in Brazil with data on climate, soils, and geology, emphasizing the important role of moisture and temperature. His textbook(the first) on ecological plant geography, Plantesainfund, published in 1895, created new terms for wetland plants (hydrophytes), moist habitat plants (mesophytes), and dry habitat plants (xerophytes). He emphasized dominant and subdominant plants of various vegetation types as a matter of convenience for discussion, but he criticized the notion that “causes” could be applied to such entities (McIntosh 1985). He described the effects of fire and succession and clearly emphasized individual plant species and many gradual gradients of vegetation, soil, and moisture. He did not view succession as necessarily unidirectional. In fact, Warming wrote that plant communities were never stable, not in equilibrium, and frequently changing due to interactions with animals, fungi, and competition among plant species (McIntosh 1985). Warming also noted that competition among species could force a species to grow in a lessprefered


History and Background, Baggage and Direction

habitat. This could lead one to misinterpret that a species “prefers” a given soil, when it was really displaced by competition in optimal locations. This finding warned future plant ecologists that species-environment relationships might include complex biotic and abiotic components and interactions. Jozef Paczoski (1864–1941), perhaps the first phytosociologist, realized that plants both respond to and create their own microenvironments in competition with other species. He discussed the interrelationships of species in plant communities. Paczoski's major contribution, although it was overlooked by most plant ecologists, was the concept of gradients between plant communities. Vegetation boundaries are often continuums, affected by microenvironments, disturbance, and various patterns of succession (Barbour et al. 1987). Leonid Ramensky (1884–1953) fully developed “the individualist concept” before Henry Gleason (Barbour et al. 1987; see below). He demonstrated before Josias Braun-Blanquet (see below) that species independently followed environmental gradients and created tables of species by foliar cover. Ramensky also developed terms for various plant strategies similar to competitors, stress-tolerators, and ruderals [before Grime (1977)], and similar to r and K strategies [before Pianka (1980); see Barbour et al. 1987]. Ramensky (1924) made several other contributions to vegetation sampling. First, he championed the need for a “quantitative and methodologically substantiated registration of facts” to support plant ecology. Second, he proposed and used a standardized square quadrat (with a diagonal of 2 m) for recording the “horizontal projection of ground shoots” (now termed foliar cover). Third, he used the standard deviation to compare variation among plots. Lastly, Ramensky ingeniously noted where plant species were absent in a plant association—a concept that would later be used to help predict probable distributions of rare plant species at landscape scales. Clinton Hart Merriam (1855–1942) formalized the concept of elevation zones of montane vegetation. His detailed descriptions of the Cascade Mountains of Washington, which focused on temperature (growing season) as the major determinant of vegetation life zones, led to his life zone maps of North America. The focus then returned to the geography of dominant species. Merriam's influence on vegetation can be seen in Bob Bailey's ecoregional maps today, and the concept of biomes ( Frederick Edward Clements (1874–1945) is the most well-known historical figure in American plant ecology (Barbour et al. 1987; McIntosh 1985; Tobey 1981). Clements drew heavily on the plant geography views of Oscar Drude (1902), who wrote about the communal life of species that caused an ever-changing dynamic factor in changing the face of the world (McIntosh 1985). Clements expounded on these views to insist on the organism-like “plant associations,” “regional formations” of vegetation, and orderly processions of succession from early “seral stages” to “climax.” He developed a broad background in plant geography by describing the distribution of every

History and Background, Baggage and Direction


plant species in Nebraska. His study sites ranged from southwestern desserts, Death Valley, and coastal California, to Pike's Peak and alpine regions in Colorado. He knew that some species were environmental indicators for documenting succession and stressful sites. Clements (1916, 1936) was dogmatic about his ideas on plant associations and succession. He would often classify discrete plant associations based on one or two dominant species. However, he recognized “open” and “closed” communities, where changes in species composition were likely or less likely, depending on the availability of limiting resources such as light, water, minerals, and air, and the use of those resources by a variety of plant species. His views on plant associations as holistic superorganisms were widely accepted in his time. The concept of stable, self-sustaining communities controlled by climate are the basis for “potential vegetation maps” which are widely used today in climate modeling, ecosystem delineation, and biome mapping. With his colleague Roscoe Pound, Clements began standardizing the area of vegetation sampling by quantifying species composition in 1 square mile areas in Nebraska (Pound and Clements 1898a,b). “Frequency” was documented: the percentage of square mile areas occupied by given species. Then, within 5 m × 5 m quadrats, they enumerated the abundance of individuals of each species. Quantitative, multiscale vegetation sampling was born. Clements later reduced the size of the small sampling unit to 1 m2 and the size of the larger sampling unit to 16 m2, greatly reducing the costs of sampling (Weaver and Clements 1938). Pound and Clements also created an index of abundance (A) that combined information on frequency, estimated cover, and abundance in quadrats (or smaller units). The index of abundance, A = (t × e × a)/T, where t is the number of sample units that contain a given species, e is the estimated mean cover of the species throughout the sample area, a is the number of individuals in quadrats, and T is the total number of sample units. Since e, the mean cover of the species throughout the sample area, was often a pure guess, and frequency was affected by quadrat size, the abundance index was criticized by many and not used often, even by Clements. However, the concept of combining information on species abundance at local scales with occurrence at landscape scales was a very clever addition to quantitative plant ecology. Clements had the most profound effects on the way vegetation is sampled today. He contributed five indelible staples to the measurement of plant diversity. First, Clements, like Ramensky (and Gleason, see below), insisted on the use of a “quadrat” in vegetation sampling. Having a standard, repeatable, statistical measure in various plant communities was revolutionary in terrestrial plant ecology. Clements stated that experience and simple observations without enumeration was invalid (McIntosh 1985). Second, viewing the world as largely homogeneous vegetation units separated by other distinct, homogeneous units, Clements guided the use of quadrat sampling


History and Background, Baggage and Direction

into recognizably homogeneous “representative” communities and formations. In this way, order could be detected from the chaotic approaches and descriptions of others in the past. Third, he emphasized an inductive procedure to test multiple working hypotheses—primarily with quantitative sampling. He used quadrats to describe current conditions (species richness and composition, frequency, cover), experiments (removal and regrowth experiments), and monitoring (assessing change over time). Fourth, the concept of an index of abundance, or combining information across spatial scales and vegetation characteristics, set the stage for “importance values”—a way to measure the dominance of particular species at particular sites—and “diversity indices”—a way to combine species richness and abundance values. These concepts and contributions are discussed later. Fifth, Clements' new direction included studies of vegetation change rather than descriptions of the status quo. This moved the discussion from simple surveys and plant species inventories to monitoring. The hope was that this would provide a better understanding of the underlying causes of orderly community change—succession. Danish plant ecologist Christen Raunkaier is best known for life form classifications of plants such as phanerophytes, chamaephytes, cryptophytes, etc., describing plant architectures and leaf traits. More important, he strongly supported quantifying plant “formations” and published several important works between 1908 and 1934 on the use of statistics in plant ecology. Raunkaier sought to “improve upon the uncertain picture we obtain by subjective estimates of plant communities” (McIntosh 1985). He noted that all species should not be weighted equally when comparing vegetation types. Raunkaier influenced vegetation sampling in several ways. First, he advanced the use of quadrats, going beyond the use of cover classes (e.g., by Clements) by recording individuals by height class to gain insights on plant species dominance. Second, he took an experimental approach to sampling by evaluating the effects of quadrat size on species richness, always looking for new ways to quantify plant frequency and dominance. Henry Chandler Cowles (1869–1939) focused on vegetation change over time—succession—on dunes near Chicago. He discussed the classification of “vegetation societies” and “vegetation cycles.” Cowles was an “observer” and often doubted the utility and meaning of “natural experiments” (McIntosh 1985). Cowles, a contemporary of Clements, noted that succession was not a straight-line process with a fixed ending. He recognized that past conditions and processes could greatly influence current vegetation conditions. Henry Gleason (1882–1975), unappreciated in his day, is now influencing plant diversity sampling in profound ways. Gleason's paper on “The Individualistic Concept of the Plant Association” (1926) strongly challenged Clements' wellaccepted body of work. Gleason was driven by a strong desire to understand the distributions of individual species. He noted that most “communities” were in the eyes and minds of investigators. Gleason thought that plant species were clumped in distribution on the landscape, but not

History and Background, Baggage and Direction


as tightly associated with other species as they were with environmental factors. He used quadrats to test whether individuals and species were randomly distributed, finding that most were distributed in clusters or patches (Gleason 1920). This work launched decades of studies on theoretical ecology and the ideas about random versus nonrandom distributions in natural populations (Greig-Smith 1957; Hutchinson 1953; Pielou 1969). Gleason had little effect on the sampling design and methods development of early plant ecologists. So strong was the Clementsian paradigm of a landscape filled with largely homogeneous communities that most early plant ecologists were preoccupied with describing, classifying, and mapping communities (Whittaker 1962). Lost was Gleason's emphasis on the distribution of individual species. Two advances by Gleason were particularly ignored by most plant ecologists in the 20th century. First, like Raunkaier, he was well aware of the effects of quadrat size on species richness. More important, Gleason recognized the importance of quadrat sampling throughout a large area rather than confining quadrat sampling to small, “representative” areas of supposed communities. He tried, in vain, to shift the emphasis of vegetation science from dominant species and artificial communities to individual species—the primary unit of evolution as noted 75 years earlier by Darwin. Josias Braun-Blanquet (1884–1980), influenced by Kerner von Marilaun, developed methods for association measurement, classification, and nomenclature (Barbour et al. 1987). The approach is sometimes referred to as the Zurich-Montpellier, or Sigmatist, school of taxonomic natural history, in which communities are typed by characteristic species, not necessarily the dominant species (McIntosh 1985). Braun-Blanquet simplified the collection of field data. An abbreviated description of the Braun-Blanquet or relevé method is as follows [see Barbour et al. (1987) for a more complete description]. An extensive hike of the study area is completed and specific vegetation communities are selected for more detailed survey. In each community, several stands are subjectively selected and thoroughly surveyed to develop as complete a species list as possible. Although this widely varies with investigators, the most typical or representative area of a stand in each community type generally is subjectively selected for further study. Detailed information is recorded in a fairly large plot (determined by a nested quadrat technique and species area curve, or the investigator's experience; see chapter 5). In each plot, the foliar cover of each species is recorded in vegetation cover classes (&;1%, 0.95), while direct species-area curves fit the data poorly. The semilog relationship has proven to be a robust species-area curve (Shmida 1984; Stohlgren et al. 1995b).

Evaluating Species Composition Overlap Jaccard's coefficient was used to compare species composition overlap among plots and vegetation types (Krebs 1989, chapter 4). The mean Jaccard's coefficient for each vegetation type was calculated from all possible pairwise comparisons between random sets of three plots. Remember that this coefficient is sensitive to sample size, plot size, and spatial autocorrelation among plots.

Evaluating the Effects of the Minimum Mapping Unit After digitizing the high-resolution (0.02 ha minimum mapping unit) vegetation map from the aerial photography, the ELIMINATE command (with KEEPEDGE option; Arc/Info version 6.0) was used to create vegetation maps with minimum mapping units of 2 ha, 50 ha, and 100 ha. For the 50 ha and 100 ha minimum mapping unit maps, the DISSOLVE command was used to remove polygons less than the minimum mapping unit and to maintain homogeneity within the combined polygons. For each map, the area and number of polygons were calculated in each of the recognized vegetation types (STATISTICS command).

Case Study on Multiphase and Multiscale Sampling


Results, Discussion, and Lessons Learned In this 754 ha study area, 330 plant species were found in the 25 sample plots (four 0.025 ha plots and twenty-one 0.1 ha plots). By stratifying the vegetation into both large, common types (e.g., lodgepole pine, ponderosa pine, and dry meadow) and potentially important small-area types (e.g., aspen, wet meadow, and burned ponderosa pine) (Stohlgren et al. 1997b), and using unbiased plot locations, we accounted for about one-third of Rocky Mountain National Park's plant species list in just 2.2 ha of sampling area.

Species-Area Curves Vary by Vegetation Type Average species-area curves for the various vegetation types showed that the aspen type had both the steepest slope and the greatest intercept, while the burned ponderosa pine type was on the other extreme (Figure 8.5). These speciesarea relationships are only accurate over the scales of measurement (i.e., they are based on plot sizes of 1 m2, 10 m2, 100 m2, and 1000 m2), but identical field methods make comparisons valid among vegetation types. Based on these species-area relationships, one can see the importance of the species-rich aspen and wet meadow types in the study landscape. However, these average species-area relationships cannot be extended to larger areas because they do not consider species composition overlap, the extent of environmental gradients, and the patchiness of species richness at landscape scales (Figure 8.5). Clearly additional information is needed on species composition overlap within and among vegetation types at multiple scales.

Figure 8.5. Average species-area curves for sampled vegetation types in the Beaver Meadows area of Rocky Mountain National Park, Colorado. The curves are log-linear regressions of the mean number of species at 10-m2, 100-m2, and 1000-m2 scales (four plots in each type). Adapted from Stohlgren et al. 1997c.


Case Study on Multiphase and Multiscale Sampling

Species Composition Overlap Varies Within Vegetation Types Spatial variation in species composition was high in the study area. Species composition overlap between the replicate plots within a vegetation type ranged from 19.9% in the wet meadow and lodgepole types (i.e., the mean of pairwise comparisons of species lists in the 1000 m2 plots) to 47.0% in the dry meadow type (Table 8.1). The low standard errors of the means suggest that species overlap was fairly consistent within all types. Still, knowing that there is 20% or 50% overlap among plots in a vegetation type helps to quantify heterogeneity within a vegetation type. Several factors can affect the measurement of species composition overlap within a vegetation type. Plot size is an obvious factor. Small plots tend to capture common species and miss locally rare species. In fairly homogeneous habitats, increases in plot size may produce similar or increased values in species composition overlap. However, in heterogeneous types, or if greater environmental gradients are included in larger plots and additional rare microhabitats are incorporated in the larger plots, then mean species composition overlap can decrease with larger plots. Such was the case here (Table 8.1). In the dry meadow, aspen, burned ponderosa, and lodgepole pine types, the large plots contained more common, patchy, and widely distributed species, so species composition overlap increased among the larger plots. In contrast, the larger plots in the wet meadow and ponderosa pine types encountered more locally rare species, so species composition overlap decreased with increasing plot size (Table 8.1). Table 8.1. Mean species composition overlap (Jaccard's coefficient) within vegetation types based on four plots per type (smaller plots used in the burned ponderosa type). Vegetation type Dry meadow Wet meadow Aspen Ponderosa pine Burned ponderosa pine Lodgepole pine Standard errors are in parentheses.

Jaccard's coefficient 100 m2 plot data 43.5% (3.1%) 20.2% (4.0%) 22.8% (2.1%) 33.1% (2.1%) 27.5% (2.5%) 19.1% (1.6%)

1000 m2 plot data 47.0% (3.9%) 19.9% (3.0%) 25.5% (2.5%) 32.3% (2.8%) 31.8% (2.2%) 19.9% (1.7%)

Case Study on Multiphase and Multiscale Sampling


Other factors can also influence species composition overlap, including the distance between plots (Stohlgren et al. 1999b), chance plot locations in unique or species-rich habitats, and unequal sampling within or among plots. Four plots in the 750 ha test area would likely have higher mean species composition overlap than four plots in a 75,000 ha area. Likewise, with a small number of sample plots, the chance placement of a plot in a speciesrich patch or a microhabitat with a unique plant assemblage could greatly decrease mean species composition overlap in a vegetation type. There are a host of assumptions when comparing species composition overlap, including equal sampling effort among plots, equal detection of all plant species at each site, and similar average environmental gradients sampled within and among plots. As usual, caution should be used in interpreting these types of results.

Species Composition Overlap Also Varies Among Vegetation Types Species overlap varied greatly among vegetation types (Table 8.2). The combined species lists from four plots showed that the composition of the wet meadow vegetation type overlapped 27.9% with the aspen community type, but less than 17% with the other vegetation types. The cross-comparisons of the ponderosa pine, burned ponderosa pine, and lodgepole pine types had between 28% and 31% overlap, which is fairly high species overlap among the pine types. Although the wet meadow and dry meadow vegetation types were close to each other on the landscape, their species composition overlapped only 16.3%. Like species composition comparisons within vegetation types, comparisons of species composition overlap among vegetation types must be cautiously interpreted. Species composition among vegetation types is affected Table 8.2. Mean species composition overlap (Jaccard's coefficient) among vegetation types based on four plots per type (smaller plots used in the burned ponderosa type). Vegetation type

Wet meadow

Dry meadow 16% Wet meadow Aspen Ponderosa pine Burned ponderosa


Ponderosa pine

20% 28%

37% 14% 26%

Burned ponder- Lodgepole pine osa 21% 23% 11% 14% 19% 27% 31% 28% 30%

The burned ponderosa plots were 250 m2, while all other plots were 1000 m2, so their values are in italics—they cannot be directly compared to the other types, but may be an index of values if larger plots were used. Adapted from Stohlgren et al. (1997b).


Case Study on Multiphase and Multiscale Sampling

by all the factors that influence species overlap results within vegetation types. Small plots may exaggerate the differences between vegetation types. The physical and environmental distance between plots and groups of plots will affect these results. However, some general patterns can be quite revealing. The mesic and more rare vegetation types, aspen and wet meadow, are more similar to each other than each is to the more xeric pine types. It may be satisfying to many “community ecologists” that species composition overlap among vegetation types is consistently lower than species composition overlap within vegetation types. This provides some support for the notion that “plant communities” might be loosely defined by groups of species with somewhat higher affinities for specific overstory tree species, a confined set of environmental conditions, or a subset of topographic, edaphic, or climatic conditions. However, it should be realized that the degree of species composition overlap might be strongly influenced by the completeness of sampling (i.e., plot size, sampling intensity, observer expertise, etc.). Additional plots in each type could easily lead to higher species composition overlap among types, since most plant species occur in more than one vegetation type, more than one state, and more than one country. Few plant species are very narrow endemic species. Likewise, the boundaries of most vegetation types cannot be clearly delineated; many are gradients and ecoclines, providing ample opportunity for a desegregation of communities in intermediate habitats.

Vegetation Patterns Depend on the Minimum Mapping Unit Selected Describing vegetation patterns with large minimum mapping units may significantly underestimate plant community diversity and the number of polygons (e.g., habitat patches and landscape complexity) (Table 8.3). At the 0.02 ha resolution, six vegetation types in 117 polygons were recognized, including the aspen type, which was scattered in clumps throughout the landscape. With the 2 ha minimum mapping unit, five vegetation types were recognized for the study area. The aspen type was absent, the size of the burned ponderosa pine habitat was half of that recognized with the 0.02 ha minimum mapping unit, and the number of polygons delineated was three times greater than for the 0.02 ha minimum mapping unit. With the 50 ha minimum mapping unit, four vegetation types were recognized, while the 100 ha minimum mapping unit showed only three common vegetation types in three polygons (Table 8.3).

Determining How Various Vegetation Types Contribute to Plant Species Richness Patterns By incorporating species richness and uniqueness data with the aerial coverage estimates of vegetation types, it is possible to evaluate the relative

Case Study on Multiphase and Multiscale Sampling


Table 8.3. Vegetation types recognized, area of each type (main table entries), and number of polygons recorded (in parentheses) with different minimum mapping units for the same 754 ha study area. Vegetation type Lodgepole pine Ponderosa pine Dry meadow Wet meadow Burned ponderosa Aspen All types combined

Minimum mapping unit 100 ha 50 ha 132 ha 137 ha (1 polygon) (2 polygons) 281 ha 281 ha (1 polygon) (1 polygons) 341 ha 279 ha (1 polygon) (2 polygons) 56 ha (1 polygons)

754 ha (3 polygons)

754 ha (6 polygons)

2 ha 147 ha (9 polygons) 266 ha (20 polygons) 270 ha (7 polygons) 65 ha (2 polygons) 5 ha (1 polygons) 754 ha (39 polygons)

0.02 ha 141 ha (16 polygons) 270 ha (48 polygons) 261 ha (31 polygons) 63 ha (3 polygons) 10 ha (7 polygons) 9 ha (12 polygons) 754 ha (117 polygons)

Adapted from Stohlgren et al. (1997a). contribution of various vegetation types to plant species richness at landscape scales (Table 8.4). It is difficult to fully identify plant species that are truly unique to a specific vegetation type without a complete survey of all plant species from strictly distinct vegetation types. The unique-to-type species in this case (Table 8.4) is an index of “relative uniqueness” based on a given similar sample size, sampling effort, and pattern of sampling. In this case, about half the plant species recorded in the wet meadow type were not recorded in the other five vegetation types, suggesting the wet meadow type is an important contributor to plant diversity in the landscape. The importance of various vegetation types to landscape-scale plant diversity also can be seen with area-weighted measures. In this example, the aspen and burned ponderosa vegetation types contained the highest number of unique species per hectare, with 25.2 and 10.2 species/ha, respectively, observed in the plots. Considering their large combined area, the pine community types contained relatively few plant species (Table 8.3). The most telling result of this study was that the habitats with the greatest number of unique species per hectare were vegetation types that are not likely to be recognized with larger minimum mapping units (Figure 8.6). High-resolution maps may be needed to accurately depict patterns of species richness in complex landscapes. We know from the species-area curves (Figure 8.3) and data on species richness and uniqueness in the various vegetation types (Table 8.4) that, in this landscape, it was important to


Case Study on Multiphase and Multiscale Sampling

Table 8.4. Total number of plant species observed in each vegetation type, unique-to-type plant species, area of vegetation type, and number of unique-to-type species per hectare of habitat. Vegetation type

No. of species observed

No. of unique-totype species

Area of vegetation type (ha)

Dry meadow Wet meadow Aspen Ponderosa pine Burned ponderosa pine Lodgepole pine Total duplicates removed)

81 148 150 88 59

31 76 50 12 10

260.8 63.1 8.8 269.9 9.8

88 330


141.1 753.8

No. of unique species per hectare of habitat 0.4 3.3 25.2 0.6 10.2 1.6

delineate and adequately sample locally rare habitats such as the aspen and burned ponderosa vegetation types. Even a map created with a 2 ha minimum mapping unit would not detect the critical aspen stands in the area (Figure 8.6, bottom). In addition, high-resolution maps (Figure 8.6, top) help to depict the contagion or patchiness of habitat.

Effectiveness of Multiscale Sampling Techniques Multiscale techniques helped to quantify plant diversity patterns in the 754 ha study area in a few weeks of sampling. Because of the nested, large plots, the variation within and among vegetation types could be approximated with only three to five replicate plots per vegetation type (Table 8.3). Still, many of the metrics used in evaluating patterns of plant diversity at landscape scales are heavily influenced by plot size, sample size, habitat heterogeneity and patchiness, the complexity of environmental gradients underlying the patterns of habitat contagion and plant species distributions, and ecological processes such as herbivory, competition, and various disturbances. Establishing large, multiscale vegetation plots can be expensive. One plot took two ecologists 2–3 hours to complete, and longer in species-rich habitats. However, the multiscale vegetation plots worked equally well in expansive vegetation types (e.g., lodgepole pine) and small habitats (e.g., aspen, burned areas). Changing the size of the plots from 20 m × 50 m in most vegetation types to 10 m × 25 m in the burned ponderosa pine type did not radically affect major results. The plots were extremely efficient at capturing known plant species in Rocky Mountain National Park. Sampling occurred in a restricted elevation range of 2500–3000 m and did not encompass subalpine, alpine tundra, and lower elevation riparian zones. Still, the plots produced 330 plant species

Case Study on Multiphase and Multiscale Sampling


Figure 8.6. The Beaver Meadows area of Rocky Mountain National Park, Colorado, with a 0.02 minimum mapping unit and a 2.0 minimum mapping unit. (approximately one-third the number of plant species recorded in the park) in the 2.2 ha area within the plots in the 754 ha sampling area. This suggests that multiscale sampling techniques may be very efficient in conducting landscapescale plant surveys in areas with poor existing data (i.e., most natural areas) (Stohlgren et al. 1995c) and plant species distribution patterns may be highly redundant within larger landscapes in the same ecoregion. Capturing one-third the number of plant species in the 1074 km2 Rocky Mountain National Park in just 25 plots (0.025–0.1 ha) suggests that existing vegetation maps in many national parks, wildlife refuges, and other natural areas could be improved substantially. Maps that usually contain information only on overstory types could be improved with a minimal amount of field work by conducting multiscale plant surveys to quantify understory species richness, cover and distribution of nonindigenous plants, and locations of habitats with high diversity or unique species assemblages. The obvious next step is to model species richness related to environmental variables (e.g., slope, aspect, and elevation) to develop a predictive plant diversity model (see chapter 14). Because plots were precisely located with a GPS and the data were collected within a geographic information system framework, they provided resource managers with a means to monitor long-term changes in plant diversity and weedy plant invasions and to evaluate the effects of various land use practices on plant diversity. They also provide an independent dataset to assess the accuracy of present and future vegetation maps (Kalkhan et al. 1995). These or other multiscale sampling techniques could be used in any area with high success because they require a minimum amount of fieldwork. The efficiency is the result of (1) recognizing potentially important vegetation


Case Study on Multiphase and Multiscale Sampling

types before stratification (small aspen stands and wetlands in our case) (Figure 8.7); (2) unbiased selection of plot locations within vegetation types; and (3) the multiscale sampling methods and evaluations of species composition overlap within and among vegetation types. The techniques are adaptable to a wide range of vegetation types simply by adjusting the dimensions of the plot and subplots (Stohlgren et al. 1995b). For example, we used 0.025 ha plots in some small, burned forest types and in alpine tundra, but larger plots may be necessary in some systems. The most important requirements of the technique are to stratify common and rare vegetation types, select unbiased sampling sites, include four scales of sampling (e.g., the 1 m2, 10 m2, and 100 m2 subplots from each 1000 m2 plot for our area), and select an appropriate minimum mapping unit.

High-Resolution Mapping Helps in Evaluating Patterns of Plant Species Diversity The size of the minimum mapping unit greatly influences our potential understanding of plant diversity patterns in three ways. First, if the minimum mapping unit is too large, some large-area vegetation types appear larger and more contiguous, some medium-area vegetation types appear reduced or increased in landscape cover, and some small-area vegetation types are entirely undetected (Table 8.3). We expected that a vegetation map created with a 100 ha minimum mapping unit (typically used in statewide vegetation maps) would report only a few cover types in our study area. However, even with the 2 ha minimum mapping unit (used in many states and National Park Service units), the species-rich aspen type would not be recognized (Figure 8.6 and Table 8.3). The 2 ha minimum mapping unit recovered 86% of the plant species, but failed to detect the aspen vegetation type (Figure 8.6) and associated unique plant species (Table 8.4). The 2 ha minimum mapping unit produced better information on species diversity patterns than the 50 ha or 100 ha minimum mapping unit, but could have extreme repercussions in estimating wildlife diversity based on habitat availability. Peet (1981) showed that aspen stands were high in plant diversity, and we demonstrate that the aspen stands also have steeper species-area curves, greater numbers of unique species, and little species overlap with other communities (except with the wet meadow type, which also is rare on the landscape). Second, increasing the minimum mapping unit size sharply decreased the number of polygons recognized (Table 8.3). If these polygons represent patches of important wildlife habitat, then our assumptions about habitat availability and connectivity may be heavily influenced by the resolution of vegetation maps. Too large a minimum mapping unit may suggest the presence of a large, contiguous habitat that does not really exist. Alternatively, important thin corridors or small patches of habitat (i.e., riparian

Case Study on Multiphase and Multiscale Sampling


Figure 8.7. Lesson 15. For all plant diversity studies, if you miss the rare habitats, hotspots of species richness, and areas with unique plant assemblages, then you miss much of the story. (Photograph of rare vegetation types, by Lisa D. Schell, used with permission) zones, stands of aspen) may not be recognized on the vegetation map where they do exist in nature. For small mammals, amphibians, and patch-specific plants and invertebrates, these small sanctuaries could be the most important features for persistence, providing for the survival of populations and metapopulations throughout the landscape (Opdam et al. 1993). The third major effect of large minimum mapping unit size is simply that finished, brightly colored maps and geographic information system themes may lead to complacency: land managers may assume that additional research, inventories, and monitoring are not a priority. Finer resolution vegetation surveys can aid in the detection and management of rare species by identifying distinctive ecosystems. The aspen, wet meadow, and burned ponderosa vegetation types covered very small portions of the landscape, but contained a high proportion of the unique plant species. Several rare species occurred in the wet meadow: the ladies' tresses orchid (Spiranthes romanzoffiana), wood lily (Lilium philadelphicum), and white bog orchid (Limnorchis dilatata ssp. albiflora). Although these species are not on any


Case Study on Multiphase and Multiscale Sampling

federal lists, they may be considered locally rare because their habitats are small and patchily distributed. This case study teaches us that small minimum mapping units allow for the development of high-resolution maps. When combined with multiscale vegetation sampling of common and rare habitat types, this approach is very helpful for evaluating patterns of plant diversity.

9 Case Study Designing a Monitoring Program for Assessing Patterns of Plant Diversity in Forests Nationwide

The Issue Consider the difficulties in designing a monitoring program for the condition and production of the nation's forests, including changes in understory plant diversity. The methods have to be flexible enough to work equally well in a variety of forest types, yet standardized enough to allow for highly comparable data across the nation. The U.S. Forest Service's Forest Health Monitoring Program accomplished this task. It is a national program that makes annual evaluations of the condition, changes, and trends in the health of forest ecosystems in the United States (Riitters et al. 1992). The monitoring program consists of a nationwide, uniform distribution of sample plots providing a large, unbiased sample of the nation's forests (1 plot/63,942 ha). Since its inception and design tests in the early 1990s (Riitters et al. 1992), the monitoring program has focused on multiple aspects of forest ecosystems, including tree growth, regeneration, and mortality; visual crown symptoms; soil condition; contaminants in foliage; growth efficiency; vertical vegetation structure and diversity; and air pollution. From the outset, attention focused on sampling design efficiency, extensive training of field crews, quality control and assurance (Cline et al. 1995), and, combined with the Forest Inventory and Analysis Program, the forest monitoring program has been very successful in monitoring the timber resources of the nation's forests. Initially, vertical vegetation structure (i.e., the amount, arrangement, and composition of forest vegetation) was seen as one of the most promising


Case Study

indicators of biotic integrity and biodiversity (Riitters et al. 1992). There was early recognition that vegetation structure was relevant to plant species diversity (e.g., supporting the Endangered Species Act of 1973) and habitat diversity (e.g., supporting the Federal Policy and Management Act of 1976) as an early warning system to help identify the effects of environmental stress (Stapanian et al. 1993). However, due to changes in interagency support and decreased funding, primary emphasis was placed on tree resources, with lesser attention on plant species diversity and other ecological indicators. In 1997, U.S. Forest Service officials recognized the incredible opportunity to enhance the timber aspects of the program with improved, systematic, long-term monitoring of understory plant diversity. Invasive nonnative plant species have been documented in several site-specific studies in forests (Stohlgren et al. 1999a, 2000b), posing problems in fire control, maintaining wildlife habitat, and protecting native plant species (Mack et al. 2000; Westbrooks 1998).

Background and Sampling Considerations As initially measured in the Forest Health Monitoring Program, species richness and cover were recorded in 12 small (1 m2) quadrats per fully forested plot by vegetation height. Most field crews employed trained foresters rather than trained botanists, so recorders sometimes lumped species into simple groups such as grasses, forbs, and shrubs, or identified taxa only to the genus level. Some regions emphasized the importance of more complete and accurate data on understory plant species composition more than other regions. As a result, species richness by strata, species accumulation curves, and vegetation volume by strata could not always be meaningfully summarized at state, regional, and national scales. This provided an opportunity to retrofit the original Forest Health Monitoring methods for evaluating understory plant diversity for two reasons: (1) the total 12 m2 area (twelve 1 m2 quadrats) sampled was probably too small to capture a significant portion of native plant diversity at each site, and (2) the single-scale approach could be easily modified to better evaluate plant diversity in patchy and heterogeneous habitats, which are typical of most habitats. Small quadrats miss species, especially those with low frequencies of occurrence (see chapter 7). Especially in forested vegetation types, the small 1 m2 quadrats gathered only incomplete data on plant diversity and exotic species primarily because most plant species are locally rare (more than 50% of the species have less than 1% foliar cover) and because plants are never randomly distributed on a plot or landscape (Stohlgren et al. 1998d). The previous methodology was modified in two important ways. First, multiscale techniques were incorporated that increased the area surveyed in each plot (Bull et al. 1998). Second, qualified botanists were added to the

Case Study


field teams. Highly qualified botanists were needed to field-identify most of the species at each site, including many locally rare species and invasive species from Europe, Asia, Africa, South America, and Australia. In this pilot study, a random subset of plots in various forest types in different regions added an “extended survey” of plant diversity on each of four subplots (Figure 9.1). For each plot, cover by species was recorded in the twelve 1 m2 quadrats and the presence of all native and exotic plant species was recorded for each of the four 168 m2 subplots. In this chapter, the results of a pilot project are presented where colleagues and I tested the modified vegetation indicators (i.e., understory diversity and general vegetation structure) to provide better data on plant diversity in the nation's forests. This research was also designed to demonstrate the potential of multiscale plant diversity data (i.e., quadrat-, subplot-, and plot-level information) to evaluate species richness patterns (e.g., hot spots of native plant diversity, early detection of the spread of exotic plant species, and to assess the suitability of forest stands as habitat for wildlife).

Study Areas and Methods Plant structure and diversity data were collected on 111 plots across Oregon (n = 14), Washington (n = 12), Colorado (n = 33), Michigan (n = 37), and Virginia (n = 15) in conjunction with annually scheduled field sampling in those states (Figure 9.1). For the 20 plots located on the eastern slope of Colorado's Rocky Mountains, ancillary information was collected on elevation, GPS location (x, y coordinates), community type (pinyon-juniper, ponderosa pine, Gambel oak, lodgepole pine, aspen, Douglas-fir, spruce), and primary land use (timberland, woodland, or rangeland). The presence of cattle grazing was also noted, though this is not part of the information

Figure 9.1. The modified, multi-scale Forest Health Monitoring Program plot design. Foliar cover by species in two height strata was recorded by species in the three 1-m2 quadrats in each 168-m2 subplot, and the remaining area in each subplot was then surveyed for plant species not recorded in the quadrats. Four subplots comprise one plot.


Case Study

normally collected on the systematically placed plots. All plots were located and established using the standard Forest Health Monitoring methods. Plot locations were predetermined from an unbiased systematic sampling grid, which covers the United States (Stolte 1997). On each of the four subplots per plot, crews established three 1 m2 quadrats at 4.6 m on 30°, 150°, and 270° azimuths from the subplot center (Figure 9.1). Crews repeated this process for each site. Botanists on the field crews used a polyvinyl chloride (PVC) frame to delineate the 1 m2 vegetation quadrat. Botanists then identified and estimated cover for all species in stratum 1 (0–0.61 m) and stratum 2 (0.61–1.83 m), with ground variables recorded in stratum 1. For each quadrat, three data elements were recorded: (1) species identification, (2) strata, and (3) plant canopy and ground variable cover. Species codes used for analysis were the standardized U.S. Natural Resource Conservation Service (NRCS) PLANTS database codes. The height stratum of the plant was recorded, with foliar cover estimated to the nearest percent. Cover was also recorded for the following ground variables: wood, water, rock, roots, duff/litter, lichen, moss, soil, trail/road, dung, and other (trash, bones, etc.) (Bull et al. 1998). The 1 m2 quadrat frame was calibrated (painted in 10 cm sections) to improve the accuracy and precision of cover estimates (Figure 9.2). Botanists

Figure 9.2. Frame used to estimate cover by species in the 1-m2 quadrat.

Case Study


were carefully trained to estimate cover. Cover was estimated only on plants or portion of plants that were inside the quadrat frame. After completing the three quadrats on a subplot, the botanist searched the entire subplot and recorded any new species not found in the quadrats. Cover classes (5% classes) were assigned to any new species encountered in the larger plots and general comments about the subplot were recorded. Botanists repeated this process for all subplots. Unknown plants found on the subplots were collected off-plot, pressed, and were later identified or mailed to the herbarium at the University of North Carolina–Chapel Hill Biota of North America Program for species identification. At a minimum, field crews obtained the number of unique species on each plot, even if identifications could not be made to the species level because of the stage of growth at the time of collection. All statistical analyses were conducted with Systat (version 9.0; SPSS, Inc., Chicago, Illinois). Due to the small sample sizes in the pilot study, nonparametric Mann-Whitney U or Kruskal-Wallis tests were used to compare plant diversity characteristics among vegetation types, land use classes, and grazing types. Simple linear regression was used to correlate native and exotic species richness by state and for all plots combined. Kriging was used to map interpolations of exotic species richness in a portion of eastern Colorado.

Results, Discussion, and Lessons Learned There was considerable variation found in plant diversity in the five pilot states. For example, native species richness per plot ranged from 6 species in a Colorado plot to 100 species in a Virginia plot. Exotic species richness per plot ranged from zero species in many plots to 21 species per plot in Oregon. The cover of exotic species ranged from 0% to 113% in one Oregon plot that contained 21 exotic species. For the 111 plots in five states, the mean cover of native species was 37.4% (±2.9%; 1 S.E.), while the mean cover of exotic species was 3.7% (±1.2%). These and the following results based on small sample sizes may not be representative of conditions throughout vegetation types, land use classes, states, or the nation, but they demonstrate the potential of the combined Forest Health Monitoring–Forest Inventory and Analysis data once the program becomes fully operational throughout the nation.

Improved Measurement of Plant Diversity and Early Detection of Invasive Exotic Plants One major feature of the revised vegetation sampling method included searching the four 168 m2 subplots for additional plant species not previously


Case Study

recorded on the twelve 1 m2 quadrats. We found that searching the larger plots aided in the early detection of exotic plant species. For all 111 plots, sampling only the twelve 1 m2 quadrats resulted in capturing, on average, 33.5% (±3.3%) fewer exotic plant species per plot. The small quadrats missed 23.9% (±6.0%) of exotic plant species per plot in Michigan, and missed 53.1% (±10.1%) of exotic plant species per plot in Washington. When only one exotic plant species was found on a plot, the twelve 1 m2 quadrats missed the species 66% of the time. On average, the quadrats also missed 32% (±1.4%) of the native plant species on each plot. These were primarily locally rare, patchily distributed plant species. Thus searching the larger subplots significantly improved measurements of native plant diversity and the early detection of exotic plant species.

Assessing Patterns of Plant Diversity Among States Kruskal-Wallis test results identified significant differences among states in native and exotic plant species richness per plot—at least for this first set of test plots (Table 9.1). For example, Virginia had twice the native species per plot compared to Colorado and Michigan. The first set of plots in Washington had three times the exotic species richness of plots in Michigan. Again, these results are preliminary data from five states and sample sizes are small and may not be representative of a larger sample. The first set of plots in the states also varied in the cover of native and exotic plant species (Table 9.1). The initial plots in Colorado had much lower foliar cover of native species compared to the other states. Exotic species cover was consistently different among states and highly variable locally, resulting in larger relative variances (coefficients of variation ranged from 37% in Virginia to 82% in Oregon). Table 9.1. Mean native and exotic plant species and foliar cover in the initial 111 plots in five pilot states [KruskalWallis (KW) tests among states]. States

Native species

Exotic species

Colorado Michigan Virginia Oregon Washington Combined KW P

21.9 (1.5) 23.2 (1.5) 57.8 (7.1) 35.6 (4.8) 37.5 (4.3) 30.6 (1.8) 33.0 < 0.001

2.2 (0.4) 1.5 (0.3) 3.4 (0.8) 4.7 (1.5) 5.2 (1.4) 2.8 (0.3) 13.5 < 0.009

Standard in parentheses.

Total native cover (%) 16.6 (2.1) 41.2 (3.4) 37.7 (7.6) 52.2 (10.3) 65.0 (14.4) 37.4 (2.9) 33.6 < 0.001

Total nonnative cover (%) 0.8 (0.4) 4.7 (2.0) 3.2 (1.2) 9.7 (8.0) 2.6 (1.2) 3.7 (1.2) 7.1 < 0.129

No. of plots 33 37 15 14 12 111

Case Study


Assessing the Increased Area and Multiscale Approach of the Design For many plots, the number of native and exotic species was highly affected by the spatial scale (area) of the sample. For example, the exotic species richness might be fairly constant with increasing area (Oregon plot in Figure 9.3), or inconsistent with increasing area (Virginia plot in Figure 9.3). This demonstrates an important benefit of a multiscale sampling design. The plot in Virginia suggests that invasive plant species are patchily distributed, with many coexisting invasive plant species at larger spatial scales compared to the Oregon plot. Based on the species-area relationship, we would expect to encounter several more exotic plant species in an area twice the size and adjacent to the selected plot in Virginia. Plant ecologists cannot assume consistent, linear increases in species richness with increasing area.

Relationship of Native and Exotic Plant Diversity The first year of data showed that areas of high native plant diversity were consistently more heavily invaded by exotic species than areas of low native diversity (Figure 9.4). For the five pilot states combined, exotic species richness was significantly, positively correlated with native species richness (r = 0.38, P < 0.001). The plots in Colorado and Virginia had relatively strong relationships between native and exotic species richness. Plots in Oregon and Washington had weaker relationships, but still had positive slopes in the correlation between native and exotic species richness (Figure 9.3). Other factors such as distance to roads, riparian zones (Stohlgren et al. 1998d), and urban areas, etc., might improve these relationships.

Figure 9.3. Example species-area relationship for exotic species in one plot in Oregon and one plot in Virginia. The 1 m2 data points are means of the twelve 1 m2 quadrats, followed by the mean of four subplots (168 m2), followed by the grand total of exotic species in the four subplots (672 m2).


Case Study

Figure 9.4. Relationship of native species richness to exotic species richness for test plots in five pilot states (and combined). For the 111 plots in the five pilot states combined, plant species richness was significantly correlated to total foliar cover (r = 0.46, P < 0.001). Exotic species cover was significantly positively correlated to native species richness (r = 0.24, P < 0.001). The cover of exotic species was significantly, positively correlated to exotic species richness (r = 0.64, P < 0.001), indicating that the conditions favoring native species may also favor establishment of exotic species and their subsequent growth. Native and exotic species cover varied significantly by land use type (Table 9.2). Rangeland plots had low native species cover, while containing the greatest cover of exotic species. Woodland plots had the lowest native and exotic species cover compared to the other land use types (Table 9.2). Again, additional sampling is needed to confirm these patterns. Sites that were Table 9.2. Example Kruskal-Wallis test results of mean species richness and foliar cover in stratum I in 240 quadrats 1 m2 for land use types in Colorado. Primary land use Native species Timber Woodland Rangeland Timber/urban KW P

3.4 (0.2) 2.1 (0.2) 1.6 (0.4) 2.4 (0.3) 18.0 < 0.001

Standard errors in parentheses.

Nonnative species 0.2 (0.1) 0.1 (0.0) 1.6 (0.4) 0.6 (0.2) 33.8 < 0.001

Native cover (%) Nonnative cover (%) 11.7 (1.6) 0.9 (0.4) 10.4 (2.1) 0.2 (0.1) 10.1 (4.7) 6.8 (2.5) 15.8 (4.1) 2.4 (1.0) 6.4 31.9 < 0.093 < 0.001

No. of plots/ quadrats 10/120 5/60 2/24 3/36

Case Study


grazed and ungrazed by cattle could only be compared for 108 ponderosa pine/pinyon-juniper quadrats (only nine test plots) (Table 9.3) in Colorado. Ungrazed plots had significantly greater cover of exotic plant species compared to grazed plots in this small subsample. However, other environmental factors (e.g., light, water, nitrogen, soil disturbance, past land use) may have been different in the gazed and ungrazed sites, so the differences cannot be unequivocally associated only with grazing or lack of grazing. Are these methods robust enough to monitor plant diversity in the nation's forests? The uniform sampling design in the five-state pilot program effectively detected important and alarming patterns that certainly justify continued monitoring. Exotic plant species have successfully invaded the most species-rich sites (Figure 9.3), the cover of exotic plant species was positively correlated with exotic species richness, and 69.4% of the plots contained at least one exotic plant species. These patterns have also been observed in local studies of Rocky Mountain National Park, Colorado, the Central Grasslands (Stohlgren et al. 1999a), and in selected riparian zones in the Central Grass-lands (Stohlgren et al. 1998c,d) and. the Pacific Northwest (DeFerrari and Naiman 1994). Still, it is only by systematic, unbiased monitoring at the national scale that national trends can be evaluated. The synthesis of results from small-scale experiments and local case studies would be insufficient in describing these broad-scale patterns. With mandates and policies to protect native species diversity in the nation's forests, it may become more important to closely monitor hotspots of native and exotic plant species richness and cover. Only about 30% of the forested plots surveyed in the five states had no exotic plant species. Many of these plots may have had exotic species nearby in forest clearings or riparian zones, along roadways, or on rangelands. Often field crews did not record understory plant diversity in plots and subplots that were located in grasslands, shrublands, and other nonforested areas. Finding that species-rich areas are being more heavily invaded than species-poor areas (Figure 9.3) complicates exotic species control efforts in the nation's forests. More targeted chemical and biological control may be needed to battle exotic plant Table 9.3. Mann-Whitney U-test results of mean species richness and foliar cover in stratum 1 in 108 quadrats (1 m2) for grazed and ungrazed (by cattle) sites in ponderosa pine and pinyon-juniper vegetation types in Colorado. Grazing status

Native species

Grazed Ungrazed U P

1.7 (0.2) 2.5 (0.3) 1057 < 0.016

Standard errors in parentheses.

Nonnative species 0.1 (0.0) 0.3 (0.1) 1286 < 0.113

Native cover (%) Nonnative cover (%) 11.1 (2.9) 0.2 (0.1) 12.1 (2.5) 1.3 (0.6) 1340 1318 < 0.533 < 0.194

No. of plots quadrats 4/48 5/60


Case Study

invasions. In any case, the careful monitoring of vegetation types in the United States may allow for the rapid assessment of the vulnerability of various habitats to invasion by exotic species. This is especially needed to assess various land use practices and other factors such as ground disturbance by small mammals, proximity to disturbed roadsides and riparian zones, and proximity to infested urban sites, etc., as major contributors to the invasion process. The real power in regional-scale analyses will come after several years of data collection and after some plots are resampled over several years to better evaluate spatial and temporal variation with larger sample sizes in multiple regions. This U.S. Forest Service monitoring program will continue to provide the first national-scale data on the plant diversity of the nation's forests. More importantly, the spatial attributes of the data allow for immediate use in controlling invasive plant species (Figure 9.5). The spatial patterning of data can be immediately useful to land managers for the early detection of noxious weeds and the protection of rare/unique plant species and habitats, and to identify healthy forests versus those forests needing immediate management attention. Control efforts can take a

Figure 9.5. Example contour map of exotic plant species richness for plots in eastern Colorado. Kriging (SYSTAT 7.0) was used to interpolate between plots. Subplot locations are shown as asterisks except for subplots with only one exotic species, which are shown as a circle.

Case Study


Figure 9.6. Lesson 16. The hotspots of native plant diversity in the nation's forests are being invaded by non-native plant species. This story would have been missed if sampling continued to occur only in 1 m2 quadrats that missed rare and patchily distributed non-native species in two-thirds of the plots. (Photograph by author)


Case Study

two-pronged approach by focusing on small, newly established populations that are less expensive to treat, and on hotspots of invasion (i.e., large, dominating populations) where the threat of extirpation of native plant species is a concern (Figure 9.5). This national-scale monitoring program is the first, statistically sound, quality assured, multiscale assessment of native and exotic plant diversity in the nation's forests. Greatly expanded datasets will document hotspots of native plant diversity and primary areas of invasion by exotic plant species, determine the effects of land use practices on plant diversity, and aid in evaluating the condition of the nation's forests. Such information is vital to adaptive management, prescribing control efforts for invasive species, and aiding in the preservation of native biodiversity. Another important benefit of the sampling design is increased comparability of data with other plant diversity monitoring programs. As a slight modification to this design, our field crews have been establishing one 168 m2 subplot (including the three 1 m2 quadrats) or pairs of subplots as a rapid assessment technique for native and nonnative plant diversity and distributions in small habitats (see BEYOND%20NAWMA%20STANDARDS.pdf). These multiscale vegetation sampling methods are directly comparable to the multiscale vegetation sampling techniques used by other agencies. The current sampling strategy with the uniform sampling grid should be augmented with stratified ground surveys in rare and important habitats and areas of high risk of invasion by exotic species (frequently missed by any systematic sampling grid), and by developing comparable datasets in non-forested areas. Based on this small case study, the number and cover of exotic species in this study is just 8–9% of total species richness, but some subplots had up to 100% cover of exotic species. Mapped displays of data (Figure 9.5) will become increasingly important in identifying and protecting areas of high native plant diversity (Figure 9.6), controlling invasive and noxious plant species, and synthesizing species-environment data for predictive modeling.

10 Case Study Patterns of Plant Invasions in Forests and Grasslands

The Issue Plant ecologists realize that invasive nonnative plant species threaten native biodiversity because they can poison livestock, clog waterways, compete with cash crops, and degrade rangelands (Westbrooks 1998). Managers of national parks, wildlife refuges, and other natural areas are equally concerned because of the potential negative effects of nonnative plant species on native plant diversity, wildlife habitat, native pollinators, fire regimes, and nutrient cycling (D'Antonio and Vitousek 1992; Stohlgren et al. 1999a; Vitousek 1990; Westbrooks 1998). Thus there is an urgent need to rapidly assess the vulnerability of natural landscapes and specific habitats to invasion (Loope and Mueller-Dombois 1989). Systematic surveys of where nonnative species have successfully invaded are needed to guide research, control, and restoration efforts. Again, since only a small portion of any large landscape or region can be affordably surveyed, modeled information on native and nonnative plant diversity, soil characteristics, topography, and climate may be needed to guide the management of invasive species in larger, unsampled areas (Chong et al. 2001; Stohlgren et al. 1997a). In this case study we carefully considered current theories, experimental evidence, and various sampling design strategies before initiating the field studies (Stohlgren 2002; Stohlgren et al. 2002).


Case Study

Background and Sampling Considerations How might plant ecologists develop hypotheses and evaluate gaps in existing knowledge in invasion ecology? Several mathematical models suggest that areas of high species diversity should be resistant to invasion by non-native species (Case 1990; Law and Morton 1996; Post and Pimm 1983; Rummel and Roughgarden 1983; Turelli 1981). The mathematical models generally claim that colonization by nonnative species should decline in the face of many strongly interacting species, which are thought to use resources more completely. However, it is difficult to measure competition or resource availability in the field, especially at landscape scales. A few field studies and small-scale experiments have reported a negative relationship between native and nonnative species richness (e.g., Fox and Fox 1986; Tilman 1997; Figure 10.1). One small-scale experiment (Levine 2000) found a positive relationship between nonnative species success and native plant diversity in one riparian site in California. In a seedaddition experiment in mature oak savanna in Minnesota, Tilman (1997) found that invasibility correlated negatively with plant species richness (n = 60, 1 m2 plots). Many of these observations, theories, and small-scale experiments could lead plant ecologists to believe that species-rich plant communities might somehow be less vulnerable to invasion by nonnative plants than species-poor communities because there might be no available

Figure 10.1. Hypothetical relationship of native and non-native species richness in natural communities assuming similar environmental characteristics among sites and similar levels of immigration and turnover, and assuming competitive exclusion is a relatively strong force (i.e., available resources are more fully used in species-rich areas). Adapted from Tilman 1999.

Case Study


resources or unused niches in species-rich areas (Grime 1973; Herben et al. 2004; MacArthur and Wilson 1967; McNaughton 1983, 1993; Tilman 1982, 1997). However, a growing number of observational studies have demonstrated that locally not all species-rich vegetation types are immune to nonnative plant invasions. Robinson and Quinn (1988) and Robinson et al. (1995) showed that species-rich areas of annual grasslands in California were more easily invaded than species-poor areas. Timmins and Williams (1991) found that the number of weeds in New Zealand's forest and scrub reserves did not correlate with the number of native species. Recently our survey of five forest and meadow vegetation types in the Colorado Rockies and four prairie types in Colorado, Wyoming, South Dakota, and Minnesota reported more extensive nonnative plant invasions in species-rich vegetation types (Stohlgren et al. 1999a). Our initial observations were limited to nine vegetation types (four 1000 m2 study plots per type) (Figure 10.2), but they raise the possibility that at a local scale, some species-rich vegetation types could be invaded. To guide research, control, and restoration activities at landscape and regional scales, additional systematic surveys are badly needed to provide land managers

Figure 10.2. Number of native and non-native species in sets of four 1000 m2 plots in vegetation types in the Central Grasslands and Colorado Rocky Mountains. Adapted from Stohlgren et al. 1999a.


Case Study

with information on the patterns and environmental factors associated with the successful invasion of nonnative plant species. Studies of plant invasion may be affected by resolution (or grain), scale (or plot size), and extent (range of habitats studied, area of the region assessed). Small-scale, site-specific experiments have yielded contradictory results showing that species-rich areas can be either less invaded (Tilman 1999) or more invaded (Levine 2000), while global-extent studies suggest species-rich areas have been heavily invaded (Lonsdale 1999). The results of small-scale experiments and evaluations of regional floras have not been particularly useful to land managers who demand landscape-scale information on which habitats are (or may be) heavily invaded. It remains a top research priority of several land management agencies to conduct systematic surveys at multiple spatial scales (e.g., plot, landscape, and biome scales) for the early detection and management of nonnative plants. The objective of this case study was to synthesize data from several studies to greatly expand the number and spatial distribution of sampling sites to assess patterns of nonnative plant invasions relative to vegetation type characteristics, topography, level of disturbance, and soil characteristics (e.g., soil texture, nitrogen, and carbon). We also develop explanatory models based on the data from the 22 vegetation types throughout the north-central United States from smaller datasets from four previous studies (Stohlgren et al. 1997b, 1998c,d, 1999b) that used the same multiscale vegetation and soil sampling methods.

Case Study Areas Four different studies were conducted in nine areas (25 sets of four plots in 22 vegetation types) between 1995 and 1998 (Figure 10.3). At each site, four multiscale 20 m × 50 m vegetation plots were sampled as described below [see Stohlgren et al. (2002) for details].

Grassland Study Sites Study locations included shortgrass steppe at the Central Plains Experimental Range (Pawnee National Grassland, Nunn, Colorado), mixed grass prairie at the High Plains Experiment Station (Cheyenne, Wyoming), northern mixed prairie in Wind Cave National Park (Hot Springs, South Dakota), and tallgrass prairie in Pipestone National Monument (Pipestone, Minnesota) (Stohlgren et al. 1998c).

Riparian Study Sites There were four study locations: one in the shortgrass steppe at the Central Plains Experimental Range and three areas of northern mixed prairie in Wind

Case Study


Figure 10.3. Map of study sites from Stohlgren et al. (2002). Cave National Park (South Dakota), Badlands National Park (South Dakota), and Bighorn Canyon National Recreation Area (Wyoming and Montana) (Stohlgren et al. 1998d).

Rocky Mountain National Park Sites Seven vegetation types were characterized with three to five modified Whittaker plots in each type in Rocky Mountain National Park, Colorado (2500–3660 m elevation). These were lodgepole pine, aspen, ponderosa pine, wet meadow, dry meadow, mixed conifer, and alpine tundra (Stohlgren and Bachand 1997; Stohlgren et al. 1999a).

Grazing Study Sites Modified Whittaker plots were placed in grazed and long-term ungrazed (more than 12 years of continued protection) at the Charles M. Russell National Wildlife Refuge, Montana; Yellowstone/Grand Teton National Parks, Wyoming; and Wind Cave National Park, South Dakota (Stohlgren et al. 1999b).


Case Study

Case Study Methods Modified Whittaker sampling plots were established at each site (see Figures 5.8 and 7.4) (Stohlgren et al. 1995b), placed with the long axis parallel to the environmental gradient. Foliar cover for each species in the under-story and the percent bare ground were estimated to the nearest percent in the ten 1 m2 subplots, and native and nonnative plant species were noted in each 1000 m2 plot. Ancillary data recorded for each plot included UTM location from a GPS and elevation, slope, and aspect. This case study included a variety of environmental data for each vegetation plot. Disturbance has been linked to plant invasions in many areas. So all plots were given a disturbance class rating as follows: 0, long-term exclosure; 1, almost no grazing by native ungulates or livestock; 2, light grazing; 3, moderate grazing; 4, heavy grazing or mowing; and 5, recent fire (past 3 years). Five soil samples were taken at each plot and analyzed for texture, carbon, and nitrogen [see Stohlgren et al. (2002)]. Climate data were gathered from the long-term weather station nearest to each site. Climate variables included mean October to June precipitation, July to September precipitation, maximum and minimum January temperatures, and maximum and minimum July temperatures [see Stohlgren et al. (2002)]. All statistical analyses were conducted with Systat (version 7.0; SPSS, Inc., Chicago, Illinois), and P < 0.05 was used to determine significance in all tests. Analyses were conducted at two scales. At the “plot scale” (0.1 ha scale), each of the 100 plots were entered into the regressions of nonnative species richness and cover as dependent variables and other plant, soil, and topographic variables were entered as independent variables. At the “vegetation type scale,” the grand number of native and nonnative species (combined species lists from four 1000 m2 plots in each vegetation type) and the mean soil, disturbance, and topographic characteristics were used in the multiple regressions. Simple linear regressions were used at both scales to determine the relationships between nonnative and native species richness and cover, topographic variables, and soil characteristics and to predict nonnative species richness and cover as described in the previous case studies. In this case study we used path coefficient analysis (Dewey and Lu 1959) (see chapter 13 for more details) to evaluate the direct and indirect relationships of the environmental factors to nonnative species richness and cover. Again we relied on forward stepwise regression, the most widely used multiple regression model (Neter et al. 1990), to compare nonnative species richness and cover at multiple scales (plot scale and vegetation type scale) in a consistent manner. This regression model may not always result in the “best” regression model for all comparisons [see Neter et al. (1990, pp. 452–453)], but the reported relationships agreed with field observations. Path coefficient analysis simply displays the standardized partial regression coefficient (direct influence) of an environmental factor on the dependent variable, with

Case Study


significant (P < 0.05) simple correlation coefficients (indirect influences) shown among environmental variables. The residual factors from the stepwise linear regressions are not presented, but they are easily calculated as R(x) = %(1 − R2) (Dewey and Lu 1959; Stohlgren and Bachand 1997).

Results, Discussion, and Lessons Learned As might be expected by broad-scale sampling, mean species richness varied considerably among plots and vegetation types (Table 10.1), a fact underscored by the range of values recorded. Native species richness ranged from 9 species/ 0.1 ha in an ungrazed northern prairie type in the Charles M. Russell National Wildlife Refuge in Montana to 71 species/0.1 ha in a grazed northern prairie type in Wind Cave National Park, South Dakota. Nonnative species richness ranged from 0 species/0.1 ha in the high elevation alpine type in Rocky Mountain National Park, Colorado to 18 species/0.1 ha in the riparian juniper-grassland type in Bighorn Canyon National Recreation Area, Wyoming. The greatest foliar cover of nonnative species was 50.5%, recorded in desert grassland in Badlands National Park in South Dakota, followed by 43.5% in tallgrass prairie at Pipestone National Monument, in Minnesota. The wide variation in native and nonnative species richness can be combined with information on soil characteristics for a better understanding of the patterns of successful invasions. Soil characteristics also varied considerably among plots and vegetation types (Table 10.2), again highlighted by recorded ranges. Percent sand in the top 15 cm of soil, for example, ranged from only 2.4% in a northern prairie plot in the Charles M. Russell National Wildlife Refuge to 87.6% in a montane meadow plot in Montana (Yellowstone/Grand Teton National Parks). Conversely, percent clay ranged from 6.7% in an aspen plot in Rocky Mountain National Park to 89.3% in the Charles M. Russell National Wildlife Refuge. Percent total nitrogen in the soil ranged from 0.02% in the riparian juniper-grassland type to 0.8% in a wet meadow plot in Rocky Mountain National Park. Likewise, percent total carbon varied from 0.3% in the riparian juniper-grassland type to 14.2 % in a mixed conifer forest plot in Rocky Mountain National Park. How might soil characteristics affect nonnative plant invasions?

Multiple Factors are Correlated to Successful Invasions There are several general patterns of invasion at the 1000 m2 plot scale. There were significant positive relationships between nonnative species richness and native species richness, total soil nitrogen, and the silt + clay content in the soil (Table 10.3). Likewise, nonnative species cover was significantly positively correlated to those same variables at the plot scale (Table 10.3).


Case Study

Table 10.1. Vegetation characteristics in one hundred 0.1 ha plots (25 sets of four plots in 22 vegetation types) in the northcentral United States.

Native Study area Vegetation type Species richness Rocky Moun- Alpine 25.0 (3.9) tain National Aspen 50.2 (3.2) Park, Colorado Wet meadow 35.0 (8.7) Lodgepole pine 29.0 (2.9) Dry meadow 41.5 (1.7) Ponderosa pine 34.8 (3.2) Mixed conifer 22.5 (6.1) High Plains Mixed grass 35.5 (2.1) Experiment prairie Center, Wyoming Wind Cave Na- Northern 41.6 (4.9) tional Park, mixed prairie, South Dakota grazed Northern 56.5 (5.4) mixed prairie, riparian Charles M. Northern 23.0 (5.9) Russell Nation- mixed prairie al Wildlife Ref- ungrazed uge, Montana Northern 19.4 (2.8) mixed praire, grazed Yellowstone/ Montane mead- 34.0 (1.7) Grand Teton ow, ungrazed National Park, Montane mead- 34.0 (3.4) Wyoming ow, grazed Central Plains Shortgrass up- 26.0 (1.8) Experimental land (1996) Range, Colora- Shortgrass up- 29.4 (1.9) do land (1997) Shortgrass ri- 44.5 (9.6) parian (1997) Badlands Na- Desert/mixed 26.0 (2.2) tional Park, grass upland South Dakota Desert/mixed 33.0 (5.5) grass riparian

Nonnative Percent cover 74.4 (4.3) 38.7 (10.5) 79.0 (6.0) 8.0 (2.0) 44.3 (1.9) 15.6 (1.1) 48.2 (15.8) 45.9 (5.9)

Species richness 0.0 5.5 (1.8) 6.5 (1.2) 1.5 (0.3) 2.0 (0.7) 2.0 (0.7) 1.8 (1.2) 2.5 (1.0)

Percent cover 0.0 6.4 (4.2) 5.8 (2.6) 0.3 (0.2) 0.6 (0.2) 0.2 (0.1) 1.0 (1.0) 0.04 (0.04)

31.9 (4.5)

6.8 (0.5)

20.8 (4.4)

38.1 (3.4)

9.0 (0.8)

12.0 (4.8)

19.8 (4.9)

1.2 (0.5)

0.4 (0.2)

19.6 (3.2)

0.8 (0.3)

0.2 (0.1)

56.8 (6.4)

2.5 (1.4)

0.2 (0.1)

50.0 (5.8)

3.3 (0.9)

2.0 (1.0)

57.5 (4.1)

1.2 (0.4)

0.04 (0.01)

48.2 (5.3)

1.1 (0.4)

0.03 (0.01)

31.6 (5.6)

3.0 (1.1)

0.16 (0.11)

22.7 (2.0)

8.8 (1.5)

21.6 (11.4)

37.8 (10.6)

9.5 (0.6)

2.4 (0.4)

Case Study

Bighorn Canyon National Recreation Area, Wyoming, Montana Pipestone National Monument, Minnesota


Mixed grass 24.3 (1.7) upland Mixed grass ri- 29.0 (4.8) parian

20.6 (2.7)

2.8 (1.9)

1.2 (1.0)

38.0 (6.3)

10.0 (2.9)

20.9 (10.7)

Tallgrass prairie 37.5 (4.1)

57.6 (12.1)

8.8 (1.0)

20.4 (8.3)

Mean of four to eight plots (standard error) Aapted from Stohlgrcn et al. (2002).


Case Study

Table 10.2. Soil characteristics in one hundred 0.1 ha plots (25 sets of four plots in 22 vegetation types) in the northcentral United States. Study area Vegetation types Alpine Rocky Mountain National Aspen Park, Colorado Wet meadow Lodgepole pine Dry meadow Ponderosa pine Mixed conifer Mixed High Plains Ex- grass praiperimental rie Center, Wyoming Wind Cave Northern mixed National prairie, Park, South Da- grazed kota Northern mixed prairie, riparian Charles M. Northern mixed Russell prairie, unNational grazed Wildlife Refuge, Northern Montana mixed prairie, grazed Montane Yellowstone/Gr- meadow and Teton ungrazed National Montane Park, meadow Wyoming grazed

C (%)

N (%)

Sand (%)

Silt (%)

Clay (%) 20.2 (0.9) 10.2 (1.5) 21.4 (7.2) 12.7 (2.1) 12.1 (1.3) 10.9 (2.1) 18.1 (3.0) 26.0 (2.9)

Elevation (m) 3007 (63) 2704 (35) 2545 (4) 2665 (33) 2612 (42) 2625 (44) 2723 (113) 1944 (13)

Disturbance (class) 1.0 (0.0) 1.0 (0.0) 1.0 (0.0) 1.0 (0.0) 1.0 (0.0) 1.0 (0.0) 1.0 (0.0) 2.5 (1.0)

2.8 (0.5) 3.5 (0.8) 5.6 (1.9) 4.2 (0.6) 1.5 (0.5) 3.0 (0.8) 7.0 (2.5) 1.8 (0.1)

0.1 (0.02) 0.1 (0.01) 0.4 (0.2) 0.2 (0.1) 0.1 (0.03) 0.1 (0.04) 0.3 (0.1) 0.2 (0.01)

65.8 (1.9) 75.6 (1.4) 59.7 (11.4) 72.8 (0.8) 72.2 (1.1) 77.3 (3.4) 66.1 (3.5) 63.0) (1.8

14.0 (2.3) 14.3 (2.2) 18.9 (4.4) 14.5 (1.4) 15.7 (0.5) 11.8 (1.5) 15.8 (2.0) 11.0 (1.6)

3.1 (0.5)

0.3 (0.03)

23.1 (2.1)

31.9 (2.4)

45.0 (0.9)

1250 (30)

2.5 (0.5)

2.8 (0.1)

0.2 (0.03)

61.4 (2.9)

15.1 (2.9)

23.5 (4.4)

1187 (26)

2.5 (0.5)

1.8 (0.2)

0.2 (0.01)

17.1 (6.7)

25.8 (1.3)

57.2 (7.4)

865 (27)

2.0 (0.0)

1.4 (0.1)

0.2 (0.01)

19.3 (6.4)

23.1 (2.7)

57.7 (7.9)

863 (19)

3.6 (0.2)

2.7 (0.5)

0.2 (0.05)

53.7 (4.7)

25.1 (2.9)

21.2 (3.2)

2077 (63)

0.0 (0.0)

4.3 (0.7)

0.3 (0.06)

58.6 (4.7)

20.4 (3.5)

21.0 (2.4)

2063 (37)

2.5 (0.2)

Case Study Central Plains Experimental Range, Colorado (shortgrass steppe) Badlands National Park, South Dakota (desert mixed grass) Bighorn Canyon National Recreation Area, Wyoming, Montana (mixed grass) Pipestone National


Upland (1996) Upland (1997)

0.9 (0.1) 0.9 (0.1)

0.1 (0.01) 0.1 (0.01)

74.3 (2.8) 72.7 (3.4)

11.1 (1.5) 6.1 (1.4)

14.7 (1.5) 21.3 (2.2)

1644 (29) 1645 (16)

3.0 (0.4) 3.1 (0.2)

Riparian (1997) Upland

0.9 (0.2) 1.5 (0.2) 1.5 (0.05)

0.1 (0.02) 0.1 (0.01) 0.1 (0.02)

56.7 (4.2) 24.2 (5.0) 28.0 (8.3)

14.6 (1.7) 30.4 (5.2) 26.0 (2.5)

28.7 (4.8) 45.3 (8.2) 46.5 (8.6)

1620 (9) 795 (27) 783 (16)

2.3 (0.3) 1.8 (0.3) 2.3 (0.5)

2.2 (0.6) 0.8 (0.2)

0.1 (0.05) 0.1 (0.05)

53.5 (4.7) 62.1 (8.6)

9.9 (6.9) 13.4 (6.2)

36.6 (4.7) 24.5 (2.7)

1378 (58) 1161 (65)

2.0 (0.0) 3.5 (0.6)

Tallgrass 4.4 prairie (0.2) Monument, Minnesota

0.4 (0.02)

11.7 (1.2)

54.1 (0.8)

34.3 (1.7)

498 (16)

2.8 (1.0)


Upland Riparian

Mean of four to eight plots (standard error). Adapted from Stohlgren et al. (2002).


Case Study

Table 10.3. Simple linear regressions of vegetation and soil characteristics as predictors of nonnative species richness and cover for the one hundred 1000 m2 plots and the 25 sets of four plots in 22 vegetation types used in the study. Dependent vari- Coefficient t able/predictors 1000 m2 plot scale (n = 100 plots): Log10 No. nonnative spp. No. of native 0.013 5.08 spp. Log10 soil % N 1.69 2.12 Soil (% sand + % 0.003 2.00 clay) Log10 cover nonnative spp. No. of native 0.014 3.60 spp. Log10 No. non- 1.12 11.9 native spp. Soil (% sand + % 0.005 2.38 clay) Log10 soil % N 2.92 2.60 Vegetation type scale (n = 25 sets of four plots): Log10 No. nonnative spp. No. of native 0.125 2.63 spp. Soil (% sand + % 0.096 1.76 clay) Log10 cover nonnative spp. Log10 No. non- 0.08 6.80 native spp. Log10 soil % N 4.7 1.74







0.036 0.048

4.5 4.0

0.21 0.21

























In addition to soil characteristics, topographic, climatic, and geographic factors were correlated to invasion success. For the 100 plots throughout the study region, there was a significant negative relationship between elevation and nonnative species richness (log10 nonnative species richness; r = −0.32, F = 11.5, P < 0.001). Only the high-elevation alpine type in Rocky Mountain National Park had no nonnative species in the sample plots. At the plot scale, 50% of the variation in nonnative species richness was explained by native plant species richness, longitude, latitude, soil total nitrogen percentage, and mean maximum July temperature (Figure 10.4a). For the 100 plots, 60% of the variation in nonnative species cover was explained by nonnative plant species richness, elevation, and total soil nitrogen (Figure 10.4b). There was also evidence that climate and soils influenced total species richness at the 1000 m2 plot scale. About 25% of the variation in total species richness is explained by mean July maximum temperature, January minimum temperature, percent clay content in the soil (a possible surrogate for water-holding capacity), October to June precipitation, and July to September precipitation (F = 6.8, P < 0.001, df = 5 and 86). Thus vegetation plots encompassed a very broad range of vegetation types, soils, and climatic systems in the synthesis that follows.

Case Study


Figure 10.4. Path coefficient diagram of environmental factors influencing non-native species richness (a) and cover (b) for one hundred 1000-m2 plots in the north-central U.S. Direct arrows to non-native species richness or cover include standardized partial regression coefficient values, while arrows between environmental variables are simple correlation coefficients. R2 is the adjusted coefficient of multiple determination. Adapted from Stohlgren et al. 2002. Also, at the vegetation type scale (25 sets of four plots), nonnative species richness was significantly positively correlated to total native species richness and total soil nitrogen (Table 10.3). Compared to the plot-scale results, the relationship was stronger for native species richness and weaker for soil total nitrogen. Also at the vegetation type scale, 64% of the variation in nonnative species richness was explained by native plant species richness, elevation, and winter (October to June) precipitation (Figure 10.5a).


Case Study

Figure 10.5. Path coefficient diagram of environmental factors influencing non-native species richness (a) and cover (b) for the 25 sites (vegetation-type scale) in the north-central U.S. Direct arrows to non-native species richness or cover include standardized partial regression coefficient values, while arrows between environmental variables are simple correlation coefficients. Adapted from Stohlgren et al. 2002. One possible interpretation is that once several nonnative species become established, a habitat may be particularly vulnerable to greatly increased cover of invasive species or eventual invasion by a highly productive species (Figures 10.5b and 10.6). This explanation is supported by the strongly positive correlation between the nonnative species richness and foliar cover at the plot scale (r = 0.77, P < 0.001) and vegetation type scale (r = 0.83, P < 0.001) (Figure 10.4). Regardless of the specific mechanisms involved in the invasion process, the patterns above (Figures 10.4 and 10.5, and Table 10.3) indicate that several biotic and abiotic factors may be interacting to affect the successful invasion of nonnative species. Single factors explain much less variation in nonnative species richness and cover (Table 10.3) than multiple factors (Figures 10.4

Case Study


Figure 10.6. Relationship of non-native species richness and cover for the 25 vegetation types (sets of four 1000-m2 plots) in the north-central U.S. and 10.5). It is also clear that different processes may be occurring at different spatial scales (e.g., the plot scale and the vegetation type scale). There are several factors that were not measured directly, such as plant biomass, seed bank or propagule pressure, resource availability for individual species, and plant competition for specific resources. It is possible that some of the factors that were directly measured are surrogates for some of the unmeasured variables. Still, more than half of the variability in nonnative species richness and cover can be obtained with relatively few biotic and abiotic factors in the study areas.

Reexamining the Relationship of Native Species Richness and Plant Invasion In both grassland and montane biomes, species-rich sites have been heavily invaded at multiple spatial scales (Tables 10.1 and 10.3, and Figures 10.4a and 10.5a) (Stohlgren et al. 1999a). This alarming pattern may be more widespread than previously thought. A recent landscape-scale survey in arid, southwestern Utah found similar patterns, with heavy invasions in areas high in native species richness, rare habitats, and fertile soils (Stohlgren et al. 2001). Locally, speciesrich riparian zones were more heavily invaded than species-poor upland sites nearby (Table 10.1) (Planty-Tabacchi et al. 1996; Stohlgren et al. 1998d). Of greatest concern, however, was a stronger correlation between native and nonnative species richness at the vegetation type scale compared to the plot scale. Species-rich vegetation types in the north-central United States appear to be highly vulnerable to invasion by nonnative species (Table 10.2). Conversely, there was little support for theories that areas of high species diversity might resist invasion by nonnative species


Case Study

(Case 1990; Law and Morton 1996; MacArthur and Wilson 1967; Post and Pimm 1983; Rummel and Roughgarden 1983; Turelli 1981). We didn't measure seed banks or seed dispersal, which surely influence patterns of plant diversity. Propagule pressure by native or nonnative species may be disproportionately higher in certain areas, but it is difficult and impractical to measure, monitor, regulate, or manage propagule pressure at landscape scales, especially for seeds that are ubiquitously distributed by wind, large and small mammals, and insects (Stohlgren 2002). The origin, autecology, and genetic variation of the nonnative invaders may also be important (Mack et al. 2000). In most of our study sites, disturbances such as grazing by ungulates and domestic animals had minimal effects on non-native species richness. Intensive grazing by cattle for more than 100 years on the relatively species-poor shortgrass steppe of Colorado has resulted in little invasion by nonnative species (Stohlgren et al. 1999b). There were no significant differences in native and nonnative species richness and cover between long-term grazed and ungrazed plots at the Charles M. Russell National Wildlife Refuge or in Yellowstone and Grand Teton National Parks (Table 10.1) (Stohlgren et al. 1999b). Nonnative species appear to be invading and thriving in both grazed and long-term ungrazed sites in our study areas (Stohlgren et al. 1999b). These sites tended to have long evolutionary histories of grazing (Milchunas and Lauenroth 1993). Native species richness in an area is likely the result of habitat heterogeneity and available resources (Lonsdale 1999), seed supply (Coffin and Lauenroth 1989c; Tilman 1997), and many other factors such as disturbance history, land use, species migration and turnover, herbivory, competition, diseases, and pathogens (Stohlgren et al. 1999a). The attribute of “native species richness” may have little or no direct effect on invasion potential (Levine and D'Antonio 1999; Lonsdale 1999; Rejmánek 1996; Stohlgren et al. 1999a). However, this does not diminish the importance of native species richness as an indicator or predictor of habitat vulnerability to invasion (Table 10.3 and Figures 10.2 and 10.3). The simplest explanation may be that nonnative plant species thrive on the same resources (high light, nitrogen, and water) as native plant species (Stohlgren et al. 1997b, 1998d, 1999a, 2000b). While we do not fully understand the mechanisms and processes that create patterns of plant diversity, this case study points to an urgent and practical need to carefully measure native species richness at multiple scales.

Assessing the Role of Abiotic Factors in Plant Invasion It is difficult to isolate and quantify potential causal factors from observational studies. Still, the relative effects of factors might be deduced by correlations of individual factors, such as soil nitrogen or disturbance, to response variables such as the number or cover of nonnative species. Disturbance, as my

Case Study


colleagues and I classified it, showed little statistical relationship to non-native species richness (Stohlgren et al. 2002). For the one hundred 0.1 ha plots, the number of plots from low (class 0) to high disturbance (class 5) were 5, 35, 26, 17, 15, and 2 plots. Despite this fairly broad range, there were no significant correlations between disturbance class and nonnative species richness at either the plot scale (P = 0.56, F = 0.33) or the vegetation type scale (P = 0.33, F = 0.98). This is not to say that disturbance is unimportant in the invasion process, it may merely draw attention to successful invasion of non-native plant species in disturbed and undisturbed habitats (Stohlgren et al. 1999b), high variation in the biotic and abiotic factors in disturbed and undisturbed sites, or that disturbances at smaller or larger scales may influence plant diversity in different ways than we recorded in this case study (Stohlgren et al. 1999b, 2002). A simple explanation may be that many native and nonnative plant species may be well adapted to typical grassland and forest disturbances with which they evolved, such as grazing, fire, flooding, and small ground-dwelling mammals. It may be that superficial disturbances, such as removal of aboveground biomass by herbivores, are less devastating than a plow, road grader, or excavations by small mammals. Traveling to our study sites we observed invasive plants along nearly all of the roadways and edges of agricultural lands. We detected no invasive plants at high-elevation alpine sites in Rocky Mountain National Park, Colorado, probably because most Mediterranean weeds in the area cannot tolerate cold temperatures (Stohlgren et al. 2000b). However, the common dandelion (Taraxacum officinale) can be found along high-elevation road cuts and trails. Other disturbances such as large-scale fire, insect and disease outbreaks, and flooding were not assessed in this case study, and they deserve more attention in future research (Hobbs and Huenneke 1992). Likewise, activities that increase available nitrogen on a site (e.g., fire, air pollution, fertilization) may promote invasion, especially if the site is near or connected to an already infested site. Several abiotic factors are correlated to native species richness, such as soil nitrogen, elevation, and precipitation (Stohlgren et al. 2002). These factors are relatively easy to document over large areas and may greatly improve the precision and accuracy of spatial models of invasive species (Chong et al. 2001). That is, the nonnative species pool may include species that favor fertile sites. Escape from natural enemies may add to the success of nonnative species (Mack et al. 2000), but pathogens are more difficult to measure and monitor. Habitat characteristics are unquestionably important predictors of successful invasions (Table 10.3 and Figures 10.2 and 10.3), and they are relatively easier and less expensive to measure and monitor (Stohlgren et al. 1998d, 1999a, 2000b). Isolating the causes of the patterns reported here are beyond the scope of this observational study. Instead, we draw the plant ecologist's attention to the locations, habitats, and physical factors associated with the current patterns of successful invasion to aid in future control and restoration efforts (Stohlgren et al. 2000b).


Case Study

Speculating on Theory, the Role of Species Turnover, and Plant Invasion Community succession theory (Clements 1916) and competitive exclusion theory (Grime 1973) might assert that plant communities with high frequencies of co-occurring or “shared species” in confined areas may use resources more completely via niche differentiation and resource partitioning among the species. One would hypothesize that plant communities with few frequently occurring species would be more vulnerable to newly invading species, while communities with many frequently co-occurring species would sequester resources more completely, insuring against newly invading species. The data from the 25 vegetation types showed no correlation between the number of plant species shared among plots in a vegetation type and the number of nonnative species in the four plots (Figure 10.7a). The theory of competitive exclusion might assert that areas of high native plant diversity would more completely occupy available niches and thus more resources, making few resources available for newly invading species. One would hypothesize a strong negative relationship between species richness and the number of nonnative species in a vegetation type. In direct contrast, the data from the 25 vegetation types showed a significant positive correlation between the number of native plant species (in four 1000 m2 plots) and the number of nonnative plant species that have successfully invaded (Figure 10.7b). Twenty-four percent of the variation in nonnative species richness could be explained by a positive correlation with native species richness. May (1973) argued that highly diverse communities are intrinsically unstable, with some species routinely dropping in and out. One would hypothesize that communities with many locally rare (or infrequent) plant species, and presumably high turnover rates, would be more susceptible to invasion than communities with few locally rare species. In this case study, as in many other field studies, my colleagues and I observed that most species in a plot had less than 1% foliar cover (Stohlgren et al. 2002). The data from the 25 vegetation types showed a strongly significant positive correlation between the number of native locally rare plant species (in four plots) and the number of nonnative plant species that have successfully invaded (Figure 10.7c). In fact, 35% of the variation in invading species richness could be explained by the richness of low-cover and locally sparse plant species. We also observed many species in cotyledon, seedling, and mature stages within and among plots. These small, young, individual plants and scattered subpopulations may be vulnerable to high turnover of individuals and local species composition. It is easy to imagine some native species dropping out and nonnative species replacing them. Theoretically many species can coexist as a result of biogenic small-scale heterogeneity and interactions among organisms for spatially and temporally variable resources (Huston and DeAngelis 1994), but species replacements also may occur in areas of

Case Study


Figure 10.7. The number of non-native species in the 25 vegetation types (sets of four 1000-m2 plots) in the northcentral U.S., relative to the number of shared species in each set of four plots, the number of native plant species, and the number of locally rare plant species (i.e., averaging 1% cover. The “sub-dominant” designation refers to the number of species with 2-m tall, >10-cm dbh Minimum stems/ha- >2m tall, >10-cm dbh Maximum stems/ha- >2m tall, >10-cm dbh Number of plots with >2m tall, >10-cm dbh Number of plots with no >2-m tall, >10-cm dbh

Modified Whittaker 1000 Intensive plot 1000 m2 m2 8 15 1125 (320.6) 867 (180.7)

Extensive plot 1000 m2 27 941 (118.8)

1675 (549.0)

1027 (270.4)

1237 (243.4)













Adapted from Barnett and Stohlgren (2003). The Extensive plot also had disadvantages. The smaller size of the plot missed species at sample points (Table 17.2), thus limiting completeness, and captured fewer unique species than the larger modified Whittaker design (Table 17.2). With no 1 m2 subplots, species cover could not be precisely recorded in the Extensive plots. Simply collecting the number and identity of species does not provide cover estimates necessary for detecting species-specific increases or decreases. For example, assessing changes in cover of the nonnative species Poa pratensis in a particular plot and across many plots may alert managers to the site-specific spread of the nonnative species (Barnett and Stohlgren 2003). Eliminating the subplots from the Extensive design limited the measure of species richness to a single scale and provided little insight into species-area relationships. This exception limits the usefulness of the information collected at each location, especially when compared to other plot locations that describe species richness at a variety of scales. Finally, collecting less information at a specific location increased the cost of travel time and therefore the cost for the information return at a local point (Barnett and Stohlgren 2003).

Benets of Nested Intensity Designs Of the three vegetation sampling designs, the modified Whittaker plot contributed the most detailed and accurate data at just a few sample locations. The advantages and the proven ability of the modified Whittaker design (Table 17.2) (Stohlgren et al. 1998a, c, 1999a, 2000a) to quantify plant species diversity at landscape scales might prompt managers to assess landscape patterns of plant diversity using only such a large and multiscale design. Other large, nested plot designs (e.g., Keeley et al. 1995) would probably have worked equally well. If we dedicated all the funding and time available

Case Study


for this inventory effort to sampling with large, nested plots, approximately 24 plots could have been placed on the landscape. Given the ability of large plot designs to capture greater species richness, and the steep modified Whittaker species accumulation curve (Figure 17.2), 24 modified Whittaker plots may have been the best option if simply generating the largest plot-based species list was the inventory objective. However, most managers must consider entire landscapes. The increased sample of 51 plots using all three sample designs may provide a more complete picture of landscape condition and variation, especially if the data collected at the smaller and single-scale plots can be leveraged as with traditional double sampling techniques (Ahmed et al. 1983). Similar to classical double sampling techniques, we leveraged the study design by relating data collected with fast and simple designs to more detailed data. We demonstrated the power of the nested intensity design by detecting hot spots of species richness across the aspen landscape. Using a multiple regression model parameterized with data from the multiscale modified Whittaker and Intensive plots, we estimated species-area curves for each Extensive plot location. Generation of species-area curves for every location allowed each plot to be ranked and spatially mapped according to the steepness of the slope or the richness at each location. Species richness may vary across scales (Gaston 2000), so the slope of a curve is a powerful tool for detecting local hot spots of diversity (Connor and McCoy 1979; Rosenzweig 1995; Stohlgren et al. 1997b). Detecting hot spots of diversity on the landscape may be important for setting priorities for preservation of important habitats or landscapes (Margules and Pressey 2000; Myers et al. 2000), directing management activities (Mulder et al. 1999; Nusser and Goebel 1997), or locating regions or plots important for inventory and monitoring purposes (Stohlgren et al. 1997b, 2000a). There are many possibilities to leverage the effectiveness of a nested intensity design. For example, we used the nested intensity approach to evaluate the spatial distribution of successful aspen regeneration. We collected more detail on stand structure at the modified Whittaker and Intensive plots than the Extensive plots, but the quantity of pole-like stems indicative of recent regeneration was recorded in all plots. Using methods similar to the species richness predictions, we were able to estimate the diameter of live and dead stems in Extensive plots. Similar manipulations could be attempted to increase the spatial understanding of aspen stand structure and regeneration for both baseline inventory and monitoring purposes. Alternatively, a manager not concerned with patchy processes such as herbivory and fire, which affect the spatial distribution of aspen regeneration (Romme et al. 1995), might simply be interested in the average number of regenerating stems on a landscape. This question may not require a large sample size facilitated by the nested intensity design, as Monte Carlo simulations suggested that only 10–15 plots would be needed to indicate that there was an average of approximately 1300 regenerating stems per hectare on this landscape. The


Case Study

ability of nested intensity techniques to evaluate the condition of aspen demonstrates that such a system can be used to address a variety of species, but may not be necessary for particular species in all cases.

Assessing Cumulative Species in an Area: Lessons Learned Examining species accumulation curves allowed us to further evaluate the contribution of each sampling design to the nested intensity design. The steeper curve of the modified Whittaker design (Figure 17.2) indicated these plots accumulate species at a faster rate than the other designs and reflects the ability of the larger modified Whittaker plot to detect more species per plot than the smaller designs (Table 17.2). The Intensive plots produced a species accumulation curve steeper than the Extensive plots (Figure 17.2). This difference must be due to the greater species diversity at some of the locations where the Intensive plots were sampled, as the Intensive and Extensive plots were the same size. We expect the trajectory of these curves would become quite similar with continued sampling that would extend the curves and dampen the effect of hot spots of diversity on the landscape. The attributes of a plot can't be assessed by plot size alone. The contribution of each plot design can be compared by quantity and the quality or usefulness of information returned for the total effort of sampling with that design. The Extensive plots do not contain subplots for collecting estimates of species cover essential for tracking information such as the spread of exotic species or the progress of restoration efforts. This omission makes the Extensive plots quicker to sample, but less useful to managers. The modified Whittaker and Intensive plots must be used if managers require cover estimates of the species detected. Sampling the modified Whittaker and Intensive plots required similar effort, approximating a ratio that allows fewer than two Intensive plots to be placed on the landscape for every one modified Whittaker plot (Table 17.1). With similar effort, the eight modified Whitaker plots provided greater species richness and cover for more species than the 15 Intensive plots (Table 17.1 and 17.2). The quality and quantity of information provided by the modified Whitaker plots dispels the popular notion that many small plots are more effective than a few large plots, at least for plant diversity studies such as this one. The species accumulation curves that describe the contribution of each plot type to the total species detected elucidates the bias of the smaller plot sizes in terms of taxonomic completeness (Figure 17.3). The flattening of the sections of the curve contributed by both the Intensive and Extensive plots suggests further sampling would detect few new species. However, regardless of the order included, the steep curve defined by the modified Whittaker plots indicates these large plots detect species missed by the smaller plots (Figure 17.3). Small plots underestimate species richness. In fact, if a manager were to iteratively assess the degree of completeness of an

Case Study


inventory either with statistical techniques (see Bunge and Fitzpatrick 1993; Schreuder et al. 1999) or looking at the slope of the species accumulation curve, the inventory might be halted prematurely if sampling were only conducted with plots even of the 100 m2 size (Barnett and Stohlgren 2003).

Other Applications for Nested Intensity Designs There are many other applications of nested intensity sampling strategies to efficiently address time, funding, and spatial and temporal accuracy concerns. For example, pilot studies (Krebs 1989; Ludwig and Reynolds 1988; Reed et al. 1993) have long been recommended for the determination of sample size. The number of study plots required to account for variability may often be underestimated, as the small number of pilot plots may not sample enough locations to accurately describe the environmental variability across the study area. The use of this nested intensity design may help an investigator better understand the extent of the variation on the landscape without the cost and time associated with the inclusion of many detailed plots on the landscape. The number of plots and the proportion of the techniques used depend on the landscape, the number of vegetation types and rare habitats, specific questions of interest, and the time and money available for the work. The inventory must be an iterative process, with the number and location of plots informed and directed as the data are collected. The nested intensity design could contribute to this iterative approach. The number of species in an Extensive plot, or the predicted species-area curve for that plot, might indicate that the plot location is important, perhaps due to high species richness or because of the presence of a particular exotic species. Such an area may require monitoring efforts more frequently than areas of less interest. The information obtained from Extensive plots may spawn the establishment of a modified Whittaker or Intensive plot at that location in future monitoring efforts to gain greater detail about species cover at finer scales and species composition at larger scales. Plant ecologists with access to computer and statistical expertise can use spatial models to predict species distributions across the landscape to continue to improve their inventory programs (see chapter 14) (Chong et al. 2001; Reich and Bravo 1998). The Extensive plots used in this design provide a means for improving and testing the accuracy of such maps without the expense of placing many new large plots on the landscape. Those ecologists who may not have adequate funding or scientific expertise to pursue such a thorough inventory system may also benefit from the nested intensity design. Simplified monitoring can be practiced between periodic intensive investigations, creating a nesting of knowledge and time. The ability to predict the number of species in a 100 m2 plot with data from 1 m2 is just one example. Those who do not have extensive taxonomy skills can simply count the number of different morphological species (based on different appearance) in Extensive plots to identify hot spots for later detailed


Case Study

research by trained taxonomists. One can learn to identify a small number of species and search Extensive plots or even the 100 m2 plot of a modified Whittaker plot. For instance, we found the easily identified and highly competitive nonnative grass Bromus inermis at several plots. If a manager learned this species and could quantify its spread in study plots, the resulting data could direct control efforts. Early detection of this species would greatly facilitate control efforts. Meanwhile, the plant ecologists could monitor just the 1 m2 subplots of a modified Whittaker or Intensive plot for the cover of such species in an effort to track subtle changes over time. An initial nested intensity inventory could direct monitoring to the highest priority areas (Stohlgren et al. 1997c). The data obtained from the abridged monitoring with 1 m2 or Extensive plots could not only direct management, but also register thresholds that suggest the need for another intensive inquiry with modified Whittaker and Intensive plots. The principles of the nested intensity inventory method described here could be adapted for use in any landscape, and to address a variety of management questions. Exploration of the methods used in this study demonstrated several benefits to inventory programs: • • • •

Nested intensity designs are well suited to document local and landscape patterns of plant diversity, while increasing time efficiency and reducing costs. A subset of fairly large plots (e.g., the modified Whittaker plot or Keeley plot) are needed to capture more complete information on species richness and unique species in some areas—methods with smaller plot sizes may miss locally rare species. A moderate set of smaller Intensive plots are needed to increase the spatial extent of the study, while increasing the accuracy and representativeness of species-area curves. A greater set of Extensive plots is needed to further increase sampling intensity in heterogeneous landscapes, thus adding unique species and locating hot spots of diversity and patchy resources of special concern.

Nested intensity systems represent creative alternatives to single-type inventory techniques commonly used by managers of both public and private lands. Newer, innovative approaches could increase the effectiveness and spatial extent of established monitoring programs.

18 Quantifying Trends in Space and Time The Issue Much of this book has focused on accurately quantifying the spatial patterns of plant diversity. This is often attempted with a combination of remote sensing data, plot sampling techniques, and interpolations from spatial models, which provide a “snapshot” of plant diversity patterns (with some quantified levels of uncertainty) (see chapter 14). However, the ultimate challenge as plant ecologists is to measure and predict (model) changes in plant diversity in space and time—more like a “movie” (Stohlgren 1999a). There are a growing number of textbooks to help us understand spatial variation (Cressie 1991; Cullinan and Thomas 1992), and there are accepted techniques for time series analysis (Pole et al. 1994; West and Harrison 1997), but there are no well-accepted textbooks on spatiotemporal analyses. The “Holy Grail” in ecological research is to develop field and modeling techniques to detect and quantify patterns in space and time and to explicate underlying mechanisms (Adler and Lauenroth 2003; Lubchenco et al. 1991). Paleoecologists and plant geographers have provided us with a basic understanding that vegetation does in fact change in space in time. Through pollen records, pack rat middens, macrofossils, and current vegetation data, they present coarse-scale descriptions of how dominant tree species and sub-dominant vegetation have “moved” across regions since the last ice age (Davis 1991; Pielou 1991). As discussed in chapter 15, the paleorecord is often spotty at best (discontinuous in time and space) and contains information on only a


Quantifying Trends in Space and Time

few species in any local flora (i.e., locally dominant species). It is possible to prove that a species was certainly “present” at some general location in the past, but it is difficult or impossible to prove that other species were truly absent. Some species leave poorer records, while others may have been hiding in small refugia or microsites nearby. Note that this situation is not unlike current vegetation sampling issues, where some plant species occur just outside the plot and are not recorded, some are in the plot as seeds and go undetected, and some may have lingered on the site (in a favorable micro-site) after establishing under different climatic conditions. This creates lag effects in temporal series of plant species distributions in response to climate change and other factors. Models of the effects of potential climate change on forest distributions have also shown that, given current speciesenvironment relationships, some vegetation types in the central Rocky Mountains will quickly migrate to Canada (Romme and Turner 1991). Such models often make simplifying assumptions that current “average” speciesenvironment relationships of a species describe the extreme behavior of individuals and various genotypes in the population and that there are no lag effects, such as individuals lingering in favorable microsites within inhospitable climate regions. Obviously such analyses are greatly limited by the grain and resolution of habitat maps, and such maps are not intended to show high resolution of all individuals within a species. Paleoecological studies and climate-vegetation models hint at what we would like to be able to do as plant ecologists: measure and monitor changes in plant diversity in space and time, usually to address general and specific questions about species-environment relationships, species persistence, and species migrations and invasions at multiple scales. I use selected questions and topical issues to intertwine with the discussion that follows: 1. A small meadow on a cattle ranch contains a population of a rare (but not federally or state-protected) lily. What would happen to the population in the long term if grazing were curtailed? 2. An invasive, nonnative plant species was recently observed just outside a large national park. Which habitat types are most vulnerable to invasion and how quickly will it likely spread? 3. Relict populations of an endemic and now rare species are showing population declines associated with tree canopy closure during natural succession. If a natural fire regime replaced the current management practice of complete fire suppression, would the rare populations flourish and spread? 4. Bountiful, colorful flowers on nonnative thistles are attracting pollinators away from less-colorful, less statuesque native plant species. How will this alter the abundance, isolation, and fecundity of the native plant species over time? 5. Hot spots of native plant diversity appear associated with rare landscape features such as wetlands, tropical forests, riparian zones, and rare forest or grassland communities. How are these

Quantifying Trends in Space and Time


hot spots threatened by land use changes, habitat loss, and fragmentation and isolation? 6. Land use changes, agriculture, and domestic livestock grazing have left only a few, small “natural” areas in many landscapes and regions—usually on unconquerable mesas or mountaintops, steep canyons, or designated protected areas. How well are most native species persisting under these pressures, and what are the long-term effects of land use on native plant diversity? 7. With greatly increased modern transportation, new plant species and genotypes, and plant pathogens from other countries are routinely entering the United States in packing materials, via horticultural trade and landscaping supplies, and by other means. How can we monitor the cumulative effects of these uninvited guests on native plant diversity?

Background and Sampling Considerations The reader should now have a strong understanding of many sampling design issues concerning the measuring and monitoring of plant diversity. Certain questions should leap to mind when presented with a question. Question 1, on the fate of the lily population in the meadow, for example, raised many additional questions. Does the timing, seasonality, or intensity of grazing directly affect the lily population in the meadow? This requires an experimental approach and perhaps greenhouse studies to substantiate alternative explanations. Does grazing interact with other factors to negatively affect the fitness (reproductive success and persistence) of the lily? How is persistence linked to genetic, species, and habitat diversity? These questions require factorial experiments, genetics research, and long-term monitoring and modeling in the meadow. Is the species successfully reproducing, or expanding or contacting it range in the meadow? This requires monitoring of a larger area, including potential, but not realized, lily habitat. Are there viable seeds in the soil? This requires soil cores, seed bank investigations, and greenhouse experiments. How rare is the species on the ranch, in the surrounding landscape, and on neighboring ranches— that is, are there other seed sources to replenish the population if it were extirpated? This requires a high-resolution survey of habitats and careful survey of lily subpopulations. How rare are the habitats in which it lives, and how plastic is the species for surviving under less-than-optimal habitats? That is, can reproduction be stimulated over broader spatial scales (larger populations generally have greater persistence than smaller populations)? This requires remote sensing work, outplanting experiments, and long-term monitoring. Are there unforeseen external threats to the lily (a new pathogen arriving, such as sudden oak death in California)? This requires vigilance and links to broader-scale research on invasive organisms and their fate.


Quantifying Trends in Space and Time

It is obvious that sampling design challenges increase with the scope of the problem. Monitoring the cover and abundance of a lily in a small meadow is far easier than landscape and regional surveys of lily habitat, detailed demographic studies of lily populations (and metapopulations), controlled yet realistic experiments on the effects of grazing and other factors on lily persistence, and population modeling over time in complex environments. Our first major challenge is in evaluating cost constraints versus the scope of the issues.

Cost Constraints versus the Scope of the Issues In the lily example above, as with all other plant diversity questions, the issue of cost cannot be ignored. The general attributes of effective vegetation monitoring efforts have been discussed by several ecologists (Krebs 1989; MacDonald et al. 1991; Risser 1993; Stohlgren 1994; Strayer et al. 1986). The attributes always include clearly articulating the goals and objectives of the studies and identifying the scope and constraints of the problem. As discussed in chapter 3, these issues must be considered within the context of appropriate sampling designs (the number, size, and pattern of sampling plots) and an evaluation of the strengths and weaknesses of alternative study designs and field techniques. The lily example also points out the need for detailed experiments and developing a strong understanding of species biology and demography, and the potential internal and external threats to population persistence. The costs of any one part of this study could be substantial, and receiving funding to conduct all aspects of the study reasonably well would be unlikely. In question 2, similar issues of scope come into play. Information is needed on the biology of the invading species; rates of spread in similar habitats; potential competitiveness of co-occurring species; the effects of current and future land use changes; the effects of manual, chemical, and biological control measures; and other factors that influence the spread and success of the invading species. As in most cases, information is needed from past research, carefully designed experiments, surveys, monitoring, and various predictive modeling approaches. Each aspect of the study is costly. As the temporal scale of the study increases, the costs of collecting detailed data for modeling may rise significantly. In question 3, fire cycles can be decades to centuries long in many forests, and careful long-term monitoring is usually beyond the scope of a single investigator. Great lengths must be taken to ensure data precision and accuracy over time, with clearly defined sampling protocols, well-marked relocatable plots, and a strong commitment to data management (Figure 18.1) (Stohlgren 1994). In most cases, long-term plant diversity studies take many measurements at relatively few sites. Most natural experiments using fire include very few true replicate burns. Thus time series analyses focus on trends in plant diversity,

Quantifying Trends in Space and Time


Figure 18.1. Combining a fully relational database linked to a geographic information system and web-based interface provides a useful way to integrate field data and modeling tools. Lesson 25. All biodiversity studies at landscape-scales and larger require extensive data management systems. Do not underestimate data management tasks and the links to surveys, monitoring, and modeling efforts. Befriend a data manager. perhaps precisely assessed but restricted spatially. Unique events and peculiar site characteristics and land use histories are fused into the analyses so temporal trends may not be extrapolated confidently to other areas (Stohlgren 1994). As the biological complexity of the issue increases, again, costs rise. In question 4, about the colorful flowers on nonnative thistles that are attracting pollinators away from native plant species, several detailed investigations are needed. Information is needed on the host specificity of the pollinators, whether pollination is inadequate for persistence at the population or species level for the native plant species, and whether the observations and study conclusions are limited to a peculiar site and time (e.g., this situation may only occur under current, temporary conditions of severe grazing or atypically small populations of native species). A costly mix of long-term monitoring, extensive surveys, controlled experiments, and the population modeling of pollinators and host plants is needed to assess changes in the abundance, isolation, fecundity, and persistence of the native plant species over time. Often basic field information is needed on the spatial distributions of multiple biological groups and environmental variables prior to designing a long-term monitoring program of plant diversity (Figure 18.2). Once basic patterns are quantified and information is available on sources of spatial and temporal variation, then integrated spatial and temporal studies can be appropriately designed. The most obvious and least appreciated challenge in addressing changes in spatial and temporal patterns of diversity is spatial scale. As in question 5,


Quantifying Trends in Space and Time

Figure 18.2. Plant ecologists must develop integrated skills in assessing the effects of multiple biological groups in monitoring spatial and temporal change in plant diversity. For example, birds and butterflies may be important in the seed dispersal, pollination, or persistence of particular plant species. the daunting task of quantifying the hot spots of native biodiversity around the globe has been attempted by a few, brave scientists, such as Meyers et al. (2000). This study often substituted expert opinion where data were lacking, and for obvious reasons. Few of the hot spots and non-hot spots of diversity have complete (or even moderately complete) species lists, and less is known about the abundance, distribution, and uniqueness of most of the species of plants and animals on our planet since fewer than a few million species of the 10–50 million species have been cataloged to date (May 1999). Thus assessing the importance of rare landscape features such as wetlands, tropical forests, riparian zones, and rare forest or grassland communities in preserving hot spots of native plant diversity is a formidable challenge (Lavoie et al. 2003). The second part of question 5 brings in various real threats to these hot spots, including land use changes, habitat loss, and fragmentation and isolation. It is clear that such an evaluation requires detailed maps and monitoring of major human activities at very local scales. What is subtle is that information is required on the intensity, frequency, and spatial patterning of the threats alone or in combination, with a detailed understanding of the effects of multiple stresses on the patterns and processes of the threats in each ecosystem (or a fair number of replicate ecosystems). Like question 5, question 6 raises issues of the sources and sinks of populations over time in a matrix of landscapes that generally includes only a small proportion of “more protected” environments in a sea of “less protected” environments (Pulliam 1988). Complex species movement patterns

Quantifying Trends in Space and Time


are difficult and costly to study, but there are some examples in the literature to guide the development of these studies (Harrison et al. 1988). The cost constraints increase as the number of factors influencing population movement increases, and the number of factors increases with the spatial and temporal scale of the study. The global nature of the cost constraints is illustrated in question 7. The magnitude and direction of invasions of plants, animals, and diseases among countries will greatly alter the way we think about measuring and monitoring native plant diversity (Dark 2004; Sax and Gaines 2003; Stohlgren et al. 2005a). Invasive diseases such as Dutch elm disease, chestnut blight, white pine blister rust, and sudden oak death have altered our forests for the foreseeable future. Invasive plant species genotypes can quickly spread and overtake native genotypes, as shown by the spread of the common reed (Phragmites australis) throughout North America (Saltonstall 2002). Wildlife and insects that depend on those trees are likely directly affected, while co-occurring species may be affected indirectly. Local environments such as undercanopy light, nutrient cycling, and fire regimes are also directly and indirectly affected by invasive species (Keeley et al. 2003; Mack et al. 2000). Question 7 speaks to the potential increased rate of future changes in plant diversity, as new pathogens are shared more frequently with greatly increased modern transportation and increasing global trade. How we monitor the cumulative effects of these uninvited guests is perhaps the greatest environmental challenge of the 21st century—and it will not be cheap or easy! For all the questions above, sampling designs must consider costs and information gain simultaneously. You often get what you pay for—so very inexpensive designs and very rapid assessments may not provide the information needed to make wise management decisions. Many other long-term monitoring designs spend too much money assessing very tiny areas of the landscape. Costs can be reduced by stratifying the landscape effectively and by using multiphase, multiscale, and nested intensity designs, as mentioned in earlier chapters. Still, cost greatly affects the size of the area surveyed or monitored, the number and size of plots, the amount of ancillary data collected, and the utility of data for modeling. Remembering that only a very small percentage of the landscape (or region, or globe) can be affordably sampled, there is an undeniable trade-off between the allocation of funds for field data collection versus modeling.

Modeling versus Data Collection Funders of plant diversity studies often begin the planning process with a statement such as, “We have x dollars allocated for this study, what can you do for us?” If a research project is awarded a limited amount of funding (as is always the case), then decisions must be made about the appropriate allocation of resources to planning, equipment, travel, field work, data management,


Quantifying Trends in Space and Time

and modeling (and publication costs if all goes well). Predictive modeling remains only a minor component of most plant ecology studies, but this is changing as the scope of the ecological questions expands (e.g., questions 1–7 above). Predictive modeling capabilities for landscape-scale and larger studies assume some expertise in remote sensing, mathematical modeling, and computer use that goes beyond the training of many plant ecologists. These skills are as useful as plant taxonomy and orienteering skills are for field ecologists, and as statistics classes and software training are for data managers. The issue here is that there are significant trade-offs when funds are allocated among the field, data management, and modeling components of most integrated studies of plant diversity. We learned from chapter 14 that the accuracy, precision, and certainty of spatial models are dependent on the sample size and placement, and on the natural variation, heterogeneity, and complexity of the landscape or region of concern. However, running different types and suites of models with the same dataset can greatly improve the amount of variation explained by the model (see Table 14.6). Thus funding either buys you more field sampling or greater modeling efforts. Other factors also come into play. Often new, improved remote sensing information becomes available after the field-work is completed. Alternatively, model outputs often show the need for additional sampling in areas of high heterogeneity and variability (i.e., where uncertainty is greater). There is no easy way to decide in advance how much funding should be allocated to field sampling versus modeling. This illustrates the need for an iterative approach to plant diversity studies. As the field study progresses, careful attention is paid to the dispersion of plots to ensure that large areas of the landscape are not missed, that the full ranges of environmental gradients are covered, and that rare habitats are not overlooked.

Theoretical and Analytical Challenges The primary theoretical and analytical challenges involve learning how to couple spatial and temporal analyses (the chicken) and develop realistic (or real) datasets that allow plant ecologists to adequately test new spatiotem-poral models (the egg). Which is needed first?

Coupling Spatial and Temporal Analyses I began the chapter by pointing out that there were several textbooks on spatial analysis (e.g., Cressie 1991) and several textbooks on time series analysis (e.g., Pole et al. 1994; Powell and Steele 1994; West and Harrison 1997), but no practical textbooks on spatiotemporal analyses. In addition, Powell and Steele (1994) pointed out that further progress on coupling spatial

Quantifying Trends in Space and Time


and temporal data in physical and biological systems is difficult to conceive without a substantial infusion of theoretical guidance (Adler and Lauenroth 2003). Many analytical tools are in the development and testing phase. For example, upscaling information from plots (or points) to landscape levels involves spatial interpolation. Yet there is disagreement whether kriging (e.g., Legendre and Fortin 1989) or a Bayesian alternative to kriging (Le and Zidek 1992) is the best geostatistical tool for the interpolation of environmental data. Where repeated measures are taken on a few trees or plots, treatment of the data requires a full consideration of the statistical issues involved (see Moser et al. 1990, Zar 1996). Haslett and Raftery (1989) provide an excellent example of space-time modeling of environmental data. They estimated the long-term average wind power output at a site in Ireland for which few data were available. The synthesis included deseasonalization, kriging, autoregressive moving average modeling, and fractional differencing in a way that can be applied directly in terrestrial ecology (e.g., to estimate spatiotemporal patterns in species richness). But these statistical models are not for novices. Modeling changes in species establishment, growth, reproduction, and spread may present a challenge to mathematical modelers. Each phase may be determined by a different suite of factors. Migration of seed to a site and establishment may be influenced by highly stochastic elements involving disturbance, wind speed and direction, or animal movements, and past and near-term precipitation and temperature patterns. Growth is influenced by the availability of resources, competition, herbivory, climate, pathogens, and many other factors. Reproduction is influenced by as many, if not more factors when you consider pollinators and their populations. The science of mathematical modeling is in its infancy relative to the age-old complexity of the patterns and processes of natural systems at multiple spatial and temporal scales (Figure 18.3). In the short-term, we might have to set low goals in modeling plant diversity. Recently colleagues have made important inroads in spatial modeling (Chong et al. 2001; Kalkhan and Stohlgren 1999). These models may be the precursor to accurate spatiotemporal models, because the first step in the process is accurately describing spatial resources. One potential approach to spatiotemporal modeling is to create “difference maps” from repeated spatial analyses. This approach might be ideal for tracking the spread and potential spread of invasive plant species. Current distributions can be mapped and modeled relative to environmental factors in the first phase. Then potential distributions or “probability of occurrence” models can be created to develop future scenarios. Finally, new field data on invading individuals and populations can be used to validate the scenarios. Implicit in this example is the need for frequent, reliable monitoring data.


Quantifying Trends in Space and Time

Figure 18.3. Lesson 26. Species-environment relationships may be different for seedlings, saplings, and mature stages of the same species, and for different genotypes within species. Slight changes in topography may drastically alter processes such as competition, herbivory, pollination, and resulting plant survivorship. Modeling species distributions, abundances, immigrations, and extirpations will require a blending of high-resolution spatial-temporal statistical models with individual-based and process-based models. (Photograph by author)

The Need for Realistic Datasets to Test Spatiotemporal Models In many cases, new datasets are needed to design and test spatiotemporal models of changes in plant diversity at landscape and larger spatial scales. Extensive field surveys and long-term monitoring must be in position to provide accurate, precise (repeatable), and fairly complete information on an annual basis. A systematic inventory of biotic resources at landscape scales will require detailed, high-resolution, remotely sensed data (Bian and Walsh 1993). Far more attention must be paid to fine-scale patterning of natural resources (Fortin et al. 1989). Test datasets are needed to develop optimum sampling strategies for monitoring changes in plant diversity in space and time (Stohlgren and Quinn 1992). Landscape scale surveys and monitoring will require combinations of systematically placed plots, stratified random sampling, targeted plots to cover extreme environmental gradients, and searching techniques to produce data that are geographically, ecologically, and taxonomically near-complete. New generations of spatiotemporal models must be developed and tested on comparable, long-term datasets to evaluate the behavior of various models on the same standardized datasets. The models must be able to “estimate”

Quantifying Trends in Space and Time


spatially and temporally variable resources and complex ecosystem processes that work at large spatial scales (e.g., Davey and Stockwell 1991; Haslett and Raftery 1989; Twery et al. 1991). Again, spatially dispersed permanent plots and long-term monitoring will be needed to validate the new models and to quantify the feedbacks between biotic and abiotic components of ecosystems and landscapes (Lubchenco et al. 1991). Finally, encompassing alternative land management strategies (Matson and Carpenter 1990) and human values and economics (National Research Council 1990) will be the key to attracting funding to obtain the necessary datasets for more model development and testing.

Institutional Challenges The institutional challenges facing future plant ecologists are substantial. Accurately quantifying long-term changes in plant diversity at landscape and larger spatial scales will take greater effort than in the past. Our past successes have been minor. The vast majority of ecological studies last only a few years, and the sizes and numbers of most vegetation plots are very small. Attracting greater funding for such endeavors is the second greatest challenge we face: a fair amount of funding has been spent on all previous plant diversity studies with only limited and usually local success. Attracting greater funding will only be successful if we take a “team approach” to plant diversity studies, incorporating new technologies and taking an experimental approach to the science of measuring plant diversity.

Taking a Team Approach to Plant Diversity Studies Early plant diversity studies remained the domain of taxonomists and naturalists typically working alone, usually by searching for and collecting plants in subjectively selected areas. This approach is ideal for developing fairly complete species lists and capturing rare species, but it is woefully lacking for quantifying changes in plant diversity over large areas through time. Later, plant ecologists added plot-based techniques, usually placed subjectively or quasi-randomly in homogeneous stands of vegetation. This approach was needed to accurately quantify changes in relatively common species, but it is limited for assessing changes in very rare species. A blending of these two approaches is needed to quantify changes in plant diversity, and taxonomists and plant ecologists can't do it alone. It is now necessary to add remote sensing specialists, spatial modelers, data managers, computer programmers, and landscape ecologists to the research team. This hypothetical example from Rocky Mountain National Park, Colorado (Figure 18.4) is based on sampling common and rare habitats with nested intensity plots (see chapter 17), long-term vegetation plots and transects


Quantifying Trends in Space and Time

Figure 18.4. Conceptual design for monitoring changes in plant diversity at landscape scales in Rocky Mountain National Park, Colorado. along elevation and moisture gradients (see chapter 16), and combining them with searching and targeted plot monitoring of very rare species populations and very rare habitats. The ultimate design is based on a blend of the theoretical approaches of Gleason, Clements, Whittaker, and others (see chapter 2). The plot sampling locations are based on stratified random sampling of dominant vegetation types in vegetation ecotones and heterogeneous areas, with additional plots in rare habitats and in sites with extreme environmental gradients. The stratification process is the responsibility of the remote sensing specialist in consultation with the landscape ecologist and statistician. The design is iterative. The samples sizes and optimal mix of plot and transect types and locations must be empirically determined and routinely tested. Sample data on spatial and temporal variation will ultimately determine sample size. Field crews contain well-trained taxonomists for immediate and accurate plant identifications (see Incorporating New Technologies below). The plots are not distributed in proportion to the area of the vegetation types, for that would oversample the common types and undersample very rare vegetation types. Sample sizes will be readjusted based on disturbance information gained and species and habitats added to

Quantifying Trends in Space and Time


the overall monitoring design. The programmer and data manager work with the statistician to provide a rule-based system for evaluating species richness and overlap, uniqueness, and patterns of species rarity as data are collected to ensure that some strata are not oversampled at the expense of missing additional rare/important strata. For example, rules would be established a priori to curtail sampling in large, homogeneous areas with few rare or unique species and sample additional areas of high diversity, high uniqueness, and high vulnerability to rapid or long-term change. The remote sensing specialists and spatial modelers have key responsibilities for initial stratification, modeling the data with levels of uncertainty, and identification of undersampled areas (and strata) for additional sampling. Everyone on the team helps to interpret and publish the findings.

Incorporating New Technologies The next generation of plant ecologists will routinely incorporate new technologies in their work if recent progress is any indication. Palmtop computers are ideal for entering data in the field and checking species lists, and for preliminary analyses to guide iterative sample efforts (Figure 18.5). A shortage of properly trained plant taxonomists may necessitate the increased development and use of species identification tools such as “polyclaves.” Polyclaves are superior to unidirectional dichotomous keys in that they allow multiple identifying characteristics to be entered simultaneously to reach a rapid solution. Combined with palmtop and satellite communications capabilities, detailed species lists, line drawings, photographs, and assistance and advice from regional taxonomic experts could be downloaded

Figure 18.5. Lesson 27. Recording data directly into palmtop computers saves time, improves the accuracy of data, and allows for immediate data analysis. Species lists, identification clues, and codes can be loaded on the palmtop in advance.


Quantifying Trends in Space and Time

to aid in species identification. GPSs have been in use for many years to help field crews find sampling locations, rare plant populations, and long-term vegetation plots. Now GPSs can be combined with video mapping technologies (e.g., Figure 18.6; to allow for quick and accurate mapping of plant populations (Stohlgren et al. 2001). Remote sensing and computer capabilities may become major expenses in future plant diversity studies. This points to the greater infrastructural support that may be required for plant diversity studies in the future, including more office and laboratory space and equipment, geographic information system and data management assistance, and modern herbarium facilities. Plant diversity studies have evolved beyond the wandering naturalist with a notebook and pocketknife-sharpened pencil. There are also institutional challenges associated with rewarding team research. Most reward systems for promotion and tenure are based on the number of senior-authored publications. There are great incentives for conducting simple, single-investigator, short-term, small-scale experiments rather than complex, multi-investigator, long-term, large-scale studies. Reward systems must adapt and recognize the growing gap between “biocomplexity” and institutional simplicity. Quantifying trends in plant diversity in space and time will be difficult and costly. The scales of studies will extend far beyond those of typical graduate degree programs in small study areas. Every effort should be made to extend the spatial and temporal scale of past and ongoing plant diversity studies, with a renewed focus on understanding the effects of increasing spatial and temporal scales on study results. This may be the only way to fully understand the link between ecological patterns and processes. The last, and greatest, institutional challenge in plant diversity studies remains—obtaining long-term commitments to research, surveys, monitoring, and modeling of plant diversity from local to global scales (Figure 18.7). Since plant species are more freely moving around the globe with modern transportation and trade (Mack et al. 2000), we can expect rapid and continuing invasions of plants, animals, and diseases. It will become increasingly difficult to protect local and distinct assemblages of native species and genotypes.

Maintaining an Experimental Approach Throughout this text I have espoused taking an experimental approach to plant diversity studies. This is especially true if the objectives of the study are to add data and information needed to understand the patterns and processes of plant diversity from local to global scales. Not all plant diversity studies will or should try to do this, but there would be significant advances in the science if data from multiple studies were comparable and complementary. Different strategies may be needed for different areas (see chapter 15), but in nearly every case, an experimental approach is likely warranted

Quantifying Trends in Space and Time


Figure 18.6. Photograph (by author) of the camcorder and GPS system tested in this study and a schematic of an application of video mapping and data recording. A global position system provides latitude, longitude, and elevation along side video from the camcorder. The data and video can be downloaded directly to a geographic information system in a computer to assess all sites with video taken. From Stohlgren et al. 2001.


Quantifying Trends in Space and Time

Figure 18.7. The future of plant diversity research is in linking patterns and processes from local to global scales. for plant diversity studies. While many different approaches can yield valuable information, I propose that many landscape-scale plant diversity studies might clearly benefit if they contain at least some of the following attributes. These are suggestions rather than “lessons.” •

Unbiased (or reduced biased) sample site selection for most plot-based studies should be complemented with purposive sampling of extreme or complex environmental gradients. Unbiased site selection allows for the generalization of results to similar habitats in the surrounding landscape. However, plot-based studies constrained by costs may preclude sampling along the full range of environmental gradients. It is important to establish some plots in extreme environments to better understand the limits of plant species distributions and patterns of diversity, uniqueness, and rarity. Plot-based surveys, monitoring, and modeling techniques, should be complemented with searching techniques. Plot-based surveys are needed to quantify patterns of plant diversity at large spatial scales with known precision and accuracy. Only a small portion of any landscape can be affordably sampled, usually less than 1%, so we

Quantifying Trends in Space and Time

• •


better be pretty good at modeling what is in the unsampled 99% of the landscape. Likewise, cost constraints limit our ability to establish plots in extremely rare habitats, so some searching techniques will always be needed to complement plot-based studies to evaluate species richness more completely. Studies should be designed to meet local objectives, but also for broad applications and synthesis with other data. There are great benefits to collecting compatible, comparable, and complementary databases on plant diversity to improve our regional and global understanding of spatial and temporal changes in plant diversity. Multiscale and nested intensity designs offer the greatest opportunities for comparable data. “Adequate replication” is a moving target and is difficult to achieve because of funding constraints and higher spatial and temporal variation. Except for very focused objectives, it is very difficult to determine appropriate sample sizes early in a study. Evaluating spatial and temporal variation is an endless task, so sampling must be an iterative process. Analysis of multiple parameters of plant diversity is the key. Not all aspects of plant diversity can be expected to change equally in space or time. Evaluating changes in species richness, cover and abundance, frequency and distribution, genetic diversity, habitat diversity, and all aspects of plant demography and evolution are important to fully understand patterns of plant diversity. Experiments aimed at isolating the underlying processes of plant diversity must be conducted in multiple vegetation types and biomes, with ranges of natural disturbances, and with many species and vegetation structures, and under a wide array of edaphic and biotic conditions. The experiments must also be complemented with surveys, monitoring, and modeling to confirm the links between patterns and processes of experimental units and natural landscapes. Monitoring genetic diversity will become increasingly important. This book has focused on plant species diversity, but the genetic diversity of many native plant species may be threatened by nonnative genotypes, hybridization, and backcrosses with native types. Nonvascular plant diversity is an important component of plant diversity, but one poorly covered by this and other texts. Standardized field sampling protocols, taxonomic experts, and automated identification keys are generally lacking for landscape and regional studies of nonnative plant diversity in many areas of the globe. More work is needed in this area. Habitat loss, land use changes, contaminants, altered disturbance regimes, invasive species, and climate change will accelerate changes in plant diversity in large regions and countries. Monitoring the rapid changes in the quality, quantity, and connectedness of habitats and species distributions remains an important component of plant diversity studies. Integrated “team science” approaches will greatly advance plant diversity studies. Quantifying patterns of plant diversity, associated


Quantifying Trends in Space and Time ecosystem processes, and associated environmental threats will require the coordinated efforts of taxonomists, remote sensing specialists, field ecologists, soil scientists, data managers, spatial modelers, and others.

In the final analysis, it will be the next generation of plant ecologists that provides us with the greatest advances in the study of plant diversity. These plant ecologists will experiment with new field techniques, remote sensing data, and computer technologies guided by more definitive theoretical frameworks. The more frequent use of multiscale sampling and nested intensity techniques, larger plots, and better replicated studies, and the collecting of detailed ancillary data integrated with improved geographic information system-based models will increase our ability to synthesize multiscale patterns of plant diversity from local to global scales. Most of our work lies ahead.

Glossary Abundance (or frequency): The relative numbers of individuals in an area (Tansley and Chipp 1926). Association: Stands of homogeneous vegetation in a community, similar to each other, yet different from other stands in the same community. Biodiversity: The variety of all life, including genetic diversity, species diversity, and habitat or ecosystem diversity. Biomass: The weight of a given individual or species in a given area (usually presented as the dry weight of living, aboveground foliage). Climax community: Those communities for which there is no evidence of replacement (Daubenmire 1968, p. 216). Community: A plant community can be understood as a combination of plants that are dependent on their environment and influence one another and modify their own environment (Mueller-Dombois and Ellenberg 1974, p. 27). Competition: When one or more plants restrict the availability of necessary resources for the growth of another plant. Connectivity: The degree to which patches of a given type are joined by corridors [adapted from Wiens (1993)]. Constancy: (1) The percentage of occurrences of a species in samples of a uniform size scattered over the geographic range of an association (Daubenmire 1968, p. 76)—it is equivalent to “frequency” among a series of stands rather than a single stand (i.e., describes a species that occurs in most (usually 50% or more) of the stands in a given vegetation type. (2) The percentage of times a species is recorded in specific plant associations growing in different areas (Tansley and Chipp 1926).



Density: The number of individuals of a species in a given area. Dominant: The most conspicuous species in an area (usually determined by basal area, cover, or density). Exclusiveness: The degree to which a given species is in one plant association and not in others (Tansley and Chipp 1926). This might also be called “uniqueness” (Stohlgren et al. 1997b). Extent: The total area under study [adapted from O'Neill et al. (1996)]. Fidelity: The degree to which a species is restricted to one community (McIntosh 1985) or the extent to which a species occurs regularly in every stand of an association. Flora: A list of species in an area (Tansley and Chipp 1926). Frequency: The presence of a species in a given number of plots or the percentage of occurrences of a species in samples of a uniform size scattered throughout a single stand. Grain: The spatial resolution of the data (O'Neill et al. 1996). Ground cover (or foliar cover): The projected cover of foliage of a species or group of species. Habitat: The sum of the effective environmental conditions under which the associations exist (Tansley and Chipp 1926). Phenology: The study of biological periodicity in relation to the seasonal sequence of climatic factors (Daubenmire 1968, p. 71). Quadrat: A small square (Pound and Clements 1898a) or rectangular (Daubenmire 1964) subplot for measuring small plant species (grasses, herbs). Resolution: The finest level of features examined in a map, photograph, or study area. Scale: The size or extent of the area being examined; the physical relationship of an area relative to a model (or map, or photograph) of the area being examined. Sere community: A successional community (e.g., pioneer species) that is later replaced by a later successional community, and ultimately the climax community. Species diversity: The number and kinds of species in an area and their genetic diversity. Species richness: The number of species in a delineated area. Vegetation: Small or large groupings of natural plants, types of forests, grasslands, etc. (Tansley and Chipp 1926).

References Adler, P. B., and W. K. Lauenroth. 2003. The power of time: spatiotemporal scaling of species diversity. Ecology Letters 6:749–756. Agee, J. K., and D. R. Johnson. 1988. Ecosystem Management for Parks and Wilderness. Seattle: University of Washington Press. Ahmed, J., C. D. Bonham, and W. A. Laycock. 1983. Comparison of techniques used for adjusting biomass estimates by double sampling. Journal of Range Management 36:217–221. Allen, R. B., R. K. Peet, and W. L. Baker. 1991. Gradient analysis of latitudinal variation in southern Rocky Mountain forests. Journal of Biogeography 18:123–139. Arrhenius, O. 1921. Species and area. Journal of Ecology 9:95–99. Ashby, E. 1948. Statistical ecology, II. A reassessment. Botanical Review 14:222–234. Austin, M. P., and P. C. Heyligers. 1991. New approach to vegetation survey design: gradsect sampling. In: C. R. Margules and M. P. Austin, eds. Nature Conservation: Cost Effective Biological Surveys and Data Analysis. Clayton South, Victoria, Australia: CSIRO; pp. 31–36. Ayres, J. M. C. 1993. As matas de várzea do Mamirauá. MCT-CNPq-Programa do trópico úmido, Sociedade civil de Mamirauá, Brazil. Backéus, I. 1993. Ecotone versus ecocline: vegetation zonation and dynamics around a small reservoir in Tanzania. Journal of Biogeography 20:209–218. Baker, W. L. 1989. Effect of scale and spatial heterogeneity on fire-interval distributions. Canadian Journal of Forest Research 19:700–706.



Baker, W. L. 1990. Species richness of Colorado riparian vegetation. Journal of Vegetation Science 1:119–124. Baker, W. L., J. A. Munroe, and A. E. Hessl. 1997. The effects of elk on aspen in the winter range of Rocky Mountain National Park. Ecography 20:155–165. Baker, W. L., and P. J. Weisberg. 1995. Landscape analysis of the forest-tundra ecotone in Rocky Mountain National Park, Colorado. Professional Geographer 47:361–375. Barbour, M. G., J. H. Burk, W. D. Pitts, F. S. Gilliam, and M. W. Schwartz. 1999. Terrestrial Plant Ecology, 3rd ed. Menlo Park, CA: Benjamin/Cummings. Barnett, D., and T. J. Stohlgren. 2001. Aspen persistence near the National Elk Refuge and Gros Ventre Valley elk feedgrounds of Wyoming, USA. Landscape Ecology 16:569–580. Barnett, D. T., and T. J. Stohlgren. 2003. A nested-intensity design for surveying plant diversity. Biodiversity and Conservation 12:255–278. Bartlein, P. J., C. Whitlock, and S. L. Shafer. 1997. Future change in the Yellowstone National Park region and its potential impact on vegetation. Conservation Biology 11:782–792. Bashkin, M., T. J. Stohlgren, Y. Otsuki, M. Lee, P. Evangelista, and J. Belnap. 2003. Soil characteristics and exotic species invasions in the Grand Stair-case-Escalante National Monument, Utah. Applied Soil Ecology 22:67–77. Bazzaz, F. A. 1975. Plant species-diversity in old-field successional ecosystems in southern Illinois. Ecology 56:485–488. Bazzaz, F. A. 1996. Plants in Changing Environments. New York: Cambridge Press. Bell, G., and M. J. Lechowicz. 1991. The ecology and genetics of fitness in forest plants, I. Environmental heterogeneity measured by explant trials. Journal of Ecology 79: 663–685. Bellehumeur, C., and P. Legendre. 1998. Multiscale sources of variation in ecological variables: modelling spatial dispersion, elaborating sampling designs. Landscape Ecology 13: 15–25. Belnap, J. 1995. Soil surface disturbances: their role in accelerating desertification. Environmental Monitoring and Assessment 37:39–57. Belnap, J. 1996. Soil surface disturbances in cold deserts: effects of nitrogenase activity in cyanobacterial-lichen soil crusts. Biology and Fertility of Soils 23:362–367. Belnap, J. 1998. Impacts of trampling soils in southeast Utah ecosystems. In: L. M. Hill, ed. Proceedings: Learning from the Land: Science in the Grand Staircase-Escalante National Monument. November 4–5, 1997. U.S. Department of the Interior, Bureau of Land Management, Southern Utah University, Cedar City, Utah. BLM Technical Report BLM/UT/GI-98/006+1220, pp. 231–244. Belnap, J., and D. A. Gillette. 1998. Vulnerability of desert soil surfaces to wind erosion: impacts of soil texture and disturbance. Journal of Arid Environments 39:133–142. Belnap, J., and K. T. Harper. 1995. Influence of cryptobiotic soil crusts on elemental content of tissue in two desert seed plants. Arid Soil Research and Rehabilitation 9:107–115.



Belsky, A. J. 1983. Small-scale patterns in grassland communities in the Serengeti National Park, Tanzania. Vegetatio 55:141–151. Belsky, A. J. 1986. Does herbivory benefit plants? A review of the evidence. American Naturalist 127:870–892. Belsky, A. J., and D. M. Blumenthal. 1997. Effects of livestock grazing on stand dynamics and soils in upland forests of the interior west. Conservation Biology 11:315–327. Berkowitz, A. R., K. Kolosa, R. H. Peters, and S. T. A. Pickett. 1989. How far in space and time can the results from a single long-term study be extrapolated? In: G. E. Likens, ed. Long-Term Studies in Ecology: Approaches and Alternatives. New York: Springer-Verlag; pp. 192–198. Bian, L., and S. J. Walsh. 1993. Scale dependencies of vegetation and topography in a mountainous environment of Montana. Professional Geography 45:1–11. Blackwood, L. G. 1991. Assurance levels of standard sample size formulas. Environmental Science and Technology 25:1366–1367. Bock, C. E., J. H. Bock, W. R. Kenney, and V. M. Hawthorne. 1984. Responses of birds, rodents, and vegetation to livestock exclosure in a semidesert grassland site. Journal of Range Management 37:239–242. Bock, C. E., J. H. Bock, and H. H. Smith. 1993. Proposal for a system of livestock exclosures on public rangelands in the western United States. Conservation Biology 7:731–733. Bock, J. H., and C. E. Bock. 1995. The challenges of grassland conservation. In: A. Joern and K. H. Keeler, eds. The Changing Prairie: North American Grasslands. New York: Oxford University Press; pp. 199–222. Bond, W. 1983. On alpha diversity and the richness of the Cape flora: a study in southern cape fynbos. In: F. J. Kruger, D. T. Mitchell, and J. U. M. Jarvis, eds. Mediterranean-Type Ecosystems: The Role of Nutrients. New York: SpringerVerlag; pp. 337–356. Bonham, C. D. 1989. Measurements for Terrestrial Vegetation. New York: John Wiley & Sons. Bonham, C. D., R. M. Reich, and K. K. Leader. 1995. A spatial cross-correlation of Bouteloua gracilis with site factors. Grasslands Science 41:196–201. Boulinier, T., J. D. Nichols, J. R. Sauer, J. E. Hines, and K. H. Pollock. 1998. Estimating species richness: the importance of heterogeneity in species detectability. Ecology 79:1018–1028. Bourdeau, P. F., and H. J. Oosting. 1959. The maritime live oak forest in North Carolina. Ecology 40:148–152. Brady, W. W., J. W. Cook, and E. F. Aldon. 1991. A microplot method for updating loop frequency range trend data: theoretical considerations and a computer simulation. Research Paper RM-295. Fort Collins, CO: Rocky Mountain Forest and Range Experiment Station, U.S. Forest Service. Bray, J. R., and J. T. Curtis. 1957. An ordination of the upland forest communities of southern Wisconsin. Ecological Monographs 27:326–349. Brooks, M. L., and J. R. Matchett. 2003. Plant community patterns in un-burned and burned blackbrush (Coleogyne ramosissima Torr.) shrublands in the Mojave Desert. Western North American Naturalist 63:283–298. Brown, B. J., and T. F. H. Allen. 1989. The importance of scale in evaluating herbivory impacts. Oikos 54:189–194.



Brown, R. L., and R. K. Peet. 2003. Diversity and invasibility of southern Appalachian plant communities. Ecology 84:32–39. Brubaker, L. B. 1986. Tree population responses to climate change. Vegetatio 67:119–130. Brubaker, L. B. 1988. Vegetation history and anticipating future vegetation change. In: J. K. Agee and D. R. Johnson, eds. Ecosystem Management for Parks and Wilderness. Seattle: University of Washington Press; pp. 41–61. Bruno, J. F., C. W. Kennedy, T. A. Rand, and M. B. Grant. 2004. Landscape-scale patterns of biological invasions in shoreline plant communities. Oikos 107:531–540. Buchholtz, C. W. 1983. Rocky Mountain National Park: A History. Boulder: Colorado Associated University Press. Bull, K. A., K. W. Stolte, and T. J. Stohlgren. 1998. Forest Health Monitoring: Vegetation Pilot Field Method Guide. Washington, DC: U.S. Forest Service. Bunge, J., and M. Fitzpatrick. 1993. Estimating the number of species: a review. Journal of American Statistical Association 88:364–373. Burkman, W. G., and G. D. Hertel. 1992. Forest Health Monitoring: a national program to detect, evaluate, and understand change. Journal of Forestry 90:26–27. Burnham, K. P., and D. R. Anderson. 2002. Model Selection and Inference: A Practical Information-Theoretic Approach. New York: Springer-Verlag. Busing, R. T., and P. S. White. 1993. Effects of area on old-growth forest attributes: implications for the equilibrium landscape concept. Landscape Ecology 8:119–126. Byers; J. E., and E. G. Noonburg. 2003. Scale dependent effects of biotic resistance to biological invasion. Ecology 84:1428–1433 Cain, S. A. 1944. Foundations of Plant Geography. New York: Hafner Press. Cambardella, C. A., and E. T. Elliott. 1993. Carbon and nitrogen distribution in aggregates from cultivated and native grassland soils. Soil Science Society of America Journal 57:1071–1076. Campbell, D. G., J. L. Stone, and A. R. Junior. 1992. A comparison of the phytosociologic and dynamics of three (varzea) forests of known ages, Rio Jurua, West Brazilian Amazon. Journal of the Linnean Society 108:213–237. Carrington, M.E. and J.E. Keeley. 1999. Comparison of postfire seedling establishment between scrub communities in mediterranean- and non–mediterranean-climate ecosystems. Journal of Ecology 87:1025–1036. Case, T. J. 1990. Invasion resistance arises in strongly interacting species-rich model competition communities. Proceedings of the National Academy of Science USA 87:9610–9614. Chaneton, E. J., and R. S. Lavado. 1996. Soil nutrients and salinity after long-term grazing exclusion in a flooding Pampa grassland. Journal of Range Management 49:182–187. Cherrill, A. J., C. McClean, P. Watson, K. Tucker, S. P. Rushton, and R. Sanderson. 1995. Predicting the distributions of plant species at the regional scale: a hierarchical matrix model. Landscape Ecology 10:197–207. Chew, R. M. 1982. Changes in herbaceous and suffrutescent perennials in



grazed and ungrazed desertified grassland in southeastern Arizona, 1958–78. American Midland Naturalist 108:159–169. Chiarucci, A., S. Maccherini, and V. De Dominicis. 2001. Evaluation and monitoring of the flora in a nature reserve by estimation methods. Biological Conservation 101:305–314. Chong, G. W. 2002. Multi-scale sampling of native and non-native plant diversity: examples of data analyses and applications. PhD dissertation. Fort Collins: Colorado State University. Chong, G. W., R. M. Reich, M. A. Kalkhan, and T. J. Stohlgren. 2001. New approaches for sampling and modeling native and exotic plant species richness. Western North American Naturalist 61:328–335. Christensen, N. L., and R. K. Peet. 1984. Convergence during secondary forest succession. Journal of Ecology 72:25–36. Cid, M. S., J. K. Detling, A. D. Whicker, and M. A. Brizuela. 1991. Vegetational responses of a mixed-grass prairie site following exclusion of prairie dogs and bison. Journal of Range Management 44:100–105. Clements, F. E. 1916. Plant Succession: An Analysis of the Development of Vegetation. Publication 242. Washington, DC: Carnegie Institution of Washington. Clements, F. E. 1936. The nature and structure of the climax. Journal of Ecology 24:252–284. Cline, S. P., S. A. Alexander, and J. E. Barnard, eds. 1995. Environmental Monitoring and Assessment Program Forest Health Monitoring Program Quality Assurance Project Plan for Detection Monitoring Project. Las Vegas, NV: U.S. Environmental Protection Agency. Cochran, W. G. 1977. Sampling Techniques, 3rd ed. New York: John Wiley & Sons. Coffin, D. P., and W. K. Lauenroth. 1989a. Disturbances and gap dynamics in a semiarid grassland: a landscape-level approach. Landscape Ecology 3:19–27. Coffin, D. P., and W. K. Lauenroth. 1989b. Small scale disturbances and successional dynamics in a shortgrass community: interactions of disturbance characteristics. Phytologia 67:258–286. Coffin, D. P., and W. K. Lauenroth. 1989c. The spatial and temporal variability in the seed bank of a semiarid grassland. American Journal of Botany 76:53–58. Cole, R. G. 2001. Staggered nested designs to estimate hierarchical levels of variability. New Zealand Journal of Marine and Freshwater Research 35: 891–896. Colebrook, J. M. 1982. Continuous plankton records: seasonal variations in the distribution and abundance of plankton in the North Atlantic Ocean and the North Sea (Calanus finmarchicus). Journal of Plankton Research 4:435–462. Collins, S. L. 1987. Interaction of disturbances in tallgrass prairie: a field experiment. Ecology 68:1243–1250. Collins, S. L., and D. E. Adams. 1983. Succession in grasslands: thirty-two years of change in a central Oklahoma tallgrass prairie. Vegetatio 51:181–190.



Collins, S. L., and S. C. Barber. 1985. Effects of disturbance on diversity in mixed-grass prairie. Vegetatio 64:87–94. Collins, S. L., and S. M. Glenn. 1991. Importance of spatial and temporal dynamics in species regional abundance and distribution. Ecology 72: 654–664. Collins, S. L., and S. M. Glenn. 1995. Grassland ecosystem and landscape dynamics. In: A. Joern and K. H. Keeler, eds. The Changing Prairie: North American Grasslands. New York: Oxford University Press; pp. 128–156. Collins, S. L., and S. M. Glenn. 1997. Effects of organismal and distance scaling on analysis of species distribution and abundance. Ecological Applications 7:543–551. Collins, S. L., S. M. Glenn, and D. J. Gibson. 1995. Experimental analysis of intermediate disturbance and initial floristic composition: decoupling cause and effect. Ecology 76:486–492. Colwell, R. K. 1997. Estimate-S: statistical estimation of species richness and shared species from samples. Version 5.0.1. Available at: Last updated September 16, 2002. Conkling, B. L., and G. E. Byers, eds. 1992. Forest Health Monitoring Field Methods Guide. Internal Report. Las Vegas, NV: U.S. Environmental Protection Agency. Connell, J. H. 1978. Diversity in tropical rain forests and coral reefs. Science 199:1302–1310. Connor, E. F., and E. D. McCoy. 1979. The statistics and biology of the species-area relationship. American Naturalist 113:791–833. Conrad, H. S. 1951. The Background of Plant Ecology. Ames: Iowa State College Press. Cook, J. W., W. W. Brady, and E. F. Aldon. 1992. Handbook for converting Parker loop frequency data to basal area. General Technical Report RM-212. Fort Collins, CO: Rocky Mountain Forest and Range Experiment Station, U.S. Forest Service. Cooper, W. S. 1913. The climax forest of Isle Royale, Lake Superior and its development. Botanical Gazette 55:1–44, 115–140, 189–235. Cornelius, J. M., and J. F. Reynolds. 1991. On determining the statistical significance of discontinuities within ordered ecological data. Ecology 72:2057–2070. Cottam, G., and J. T. Curtis. 1949. A method for making rapid surveys of woodlands by means of pairs of randomly selected trees. Ecology 30:101–104. Coughenour, M. B. 1991. Biomass and nitrogen responses to grazing of upland steppe on Yellowstone's northern winter range. Journal of Applied Ecology 28:71–82. Coughenour, M. B. 1993. Savanna—Landscape and Regional Ecosystem Model: Documentation. Fort Collins, CO: Natural Resource Ecology Laboratory, Colorado State University. Coughenour, M. B., and F. J. Singer. 1996. The concept of overgrazing and its application to Yellowstone's northern winter range. In: Technical Report NPS/NRYELL/NRTR/96–01. Denver, CO: U.S. Department of the Interior, National Park Service, Natural Resource Program Center, Natural Resource Information Division; pp. 1–11.



Coughenour, M. B., F. J. Singer, and J. J. Reardon. 1991. The Parker transects revisited: long-term herbaceous vegetation trends on Yellowstone's northern winter range. In: D. Despain, ed. Plants and Their Environments: 1st Bicentennial Science Conference on the Greater Yellowstone Ecosystem. Washington, DC: U.S. Department of the Interior, National Park Service; pp. 73–84. Crawley, M. J. 1983. Herbivory: The Dynamics of Plant-Animal Interactions. Oxford: Blackwell. Cressie, N. 1985. Fitting variogram models by weighted least squares. Mathematical Geology 17:563–586. Cressie, N. 1991. Statistics for Spatial Data. New York: John Wiley & Sons. Crumley, C. L. 1993. Analyzing historic ecotonal shifts. Ecological Applications 3:377–384. Cullinan, V. I., and J. M. Thomas. 1992. A comparison of quantitative methods for examining landscape pattern and scale. Landscape Ecology 7:221–227. Currie, D. J. 1991. Energy and large-scale patterns of animal- and plant-species richness. American Naturalist 137:27–49. Curtis, J. T., and R. P. McIntosh. 1951. An upland forest continuum in the prairie-forest border region of Wisconsin. Ecology 32:476–498. Czaplewski, R. L., and R. M. Reich. 1993. Expected Value and Variance of Moran's Bivariate Spatial Autocorrelation Statistic Under Permutation. Research Paper RM-309. Fort Collins, CO: U.S. Department of Agriculture, Rocky Mountain Experimental Range Station. Dallmeier, F. 1992. Long-Term Monitoring of Biological Diversity in Tropical Forest Areas: Methods for Establishment and Inventory of Permanent Plots. MAP Digest 11. Paris, France: United Nations Educational, Scientific, and Cultural Organization. Dallmeier, F., C. M. Taylor, J. C. Mayne, M. Kabel, and R. Rice. 1992. Effects of Hurricane Hugo on the Bisley Biodiversity Plot, Luquillo Biosphere Reserve, Puerto Rico. In: F. Dallmeier, ed. Long-Term Monitoring of Biological Diversity in Tropical Forest Areas: Methods for Establishment and Inventory of Permanent Plots. MAP Digest 11. Paris, France: United Nations Educational, Scientific, and Cultural Organization. D'Antonio, C. M., and P. M. Vitousek. 1992. Biological invasions by exotic grasses, the grass/fire cycle, and global change. Annual Review of Ecology and Systematics 23:63–87. Dark, S. J. 2004. The biogeography of invasive alien plants in California: an application of GIS and spatial regression analysis. Diversity and Distributions 10:1–9. Darwin, C. 1859. The Origin of Species. London: Penguin Books [reprint]. Daubenmire, R. 1959. A canopy-coverage method of vegetational analysis. Northwest Science 33:43–64. Daubenmire, R. F. 1940a. Exclosure technique in ecology. Ecology 21:514–515. Daubenmire, R. F. 1940b. Plant succession due to overgrazing in the Agropyron bunchgrass prairie of southeastern Washington. Ecology 21:55–64. Daubenmire, R. F. 1968. Plant Communities: A Textbook of Plant Synecology. New York: Harper and Row.



Daubenmire, R. F. 1974. Plants and Environment, 3rd ed. New York: John Wiley & Sons. Davey, S. M., and D. R. B. Stockwell. 1991. Incorporating wildlife habitat into an AI environment: concepts, theory, and practicalities. AI Applications 5:59–104. Davies, K. E., P. Chesson, S. Harrison, B. D. Inouye, B. A. Melbourne, and K. J. Rice. 2005. Spatial heterogeneity explains the scale dependence of the native-exotic diversity relationship. Ecology 86:1602–1610. Davis, M. B. 1983. Quaternary history of deciduous forest of eastern North America and Europe. Annals of the Missouri Botanical Garden 70:550–563. Davis, M. B. 1989. Retrospective studies. In: G. E. Likens, ed. Long-Term Studies in Ecology: Approaches and Alternatives. New York: Springer-Verlag; pp. 71–89. Davis, M. B. 1991. Insights from paleoecology on global change. Ecological Society of America Bulletin 70:222–228. Day, R. W., and G. P. Quinn. 1989. Comparisons of treatments after an analysis of variance in ecology. Ecological Monographs 59:433–463. DeByle, N. V., and R. P. Winokur, eds. 1985. Aspen: Ecology and Management in the Western United States. Fort Collins, CO: Rocky Mountain Forest and Range Experiment Station, U.S. Forest Service. DeFalco, L. A. 1995. Influence of Cryptobiotic Crusts on Winter Annuals and Foraging Movements of the Desert Tortoise. Fort Collins: Department of Biology, Colorado State University. DeFerrari, C. M., and R. J. Naiman. 1994. A multi-scale assessment of the occurrence of exotic plants on the Olympic Peninsula, Washington. Journal of Vegetation Science 5:247–258. Dewey, D. R., and K. H. Lu. 1959. A correlation and path-coefficient analysis of components of crested wheatgrass seed production. Agronomy Journal 51:515–518. Diaz, S., J. G. Hodgson, K. Thompson, M. Cabido, J. H. C. Cornelissen, A. Jalili, G. Montserrat-Marti, J. P. Grime, F. Zarrinkamar, Y. Asri, S. R. Band, S. Basconcelo, P. Castro-Diez, G. Funes, B. Hamzehee, M. Khoshnevi, N. PerezHarguindeguy, M. C. Perez-Rontome, F. A. Shirvany, F. Vendramini, S. Yazdani, R. Abbas-Azimi, A. Bogaard, S. Boustani, M. Charles, M. Dehghan, L. de Torres-Espuny, V. Falczuk, J. Guerrero-Campo, A. Hynd, G. Jones, E. Kowsary, F. Kazemi-Saeed, M. Maestro-Martinez, A. Romo-Diez, S. Shaw, B. Siavash, P. Villar-Salvador, and M. R. Zak. 2004. The plant traits that drive ecosystems: evidence from three continents. Journal of Vegetation Science 15:295–304. Diersing, V. E., R. B. Shaw, and D. J. Tazik. 1992. US Army land condition trend analysis (LCTA) program. Environmental Management 16:405–414. DiTommaso, A., and L. W. Aarssen. 1989. Resource manipulations in natural vegetation: a review. Vegetatio 84:9–29. Dix, R. L. 1960. The effects of burning on the mulch structure and species composition of grasslands in western North Dakota. Ecology 41:49–56. Dix, R. L., and J. E. Butler. 1954. The effects of fire on a dry, thin-soil prairie in Wisconsin. Journal of Range Management 7:265–268. Dobzhansky, T. 1950. Evolution in the tropics. American Scientist 38:209–221.



Docters van Leeuwen, W. M. 1936. Krakatau, 1883 to 1933. A. Botany. Jardin Botanique de Buitenzorg (Buitenzorg, Java), Annales vols. 46–47. Dormaar, J. F., B. W. Adams, and W. D. Willms. 1994. Effect of grazing and abandoned cultivation on a StipaBouteloua community. Journal of Range Management 47:28–32. Drude, O. 1902. Grundzüge der Pflanzenverbreitung. Vol. VI. In: A. Engler and O. Drude, eds. Die Vegetation der Erde. Leipzig: Wilhelm Engelmann; p. 671. Duffy, D. C., and A. J. Meier. 1992. Do Appalachian herbaceous understories ever recover from clearcutting? Conservation Biology 6:196–201. Dukes, J. S. 2002. Species composition and diversity affect grassland susceptibility and response to invasion. Ecological Applications 12:602–617. Ellenberg, H. 1956. Aufgaben und Methoden der Vegetationskunde. Stuttgart: Eugen Ulmer. Ellenberg, H. 1958. Bodenreaktion (einschlieblich Kaltfrage). In: W. Ruhland, ed. Handbuch der Pflanzenphysiologie. Berlin: Springer-Verlag; pp. 638–708. Elliott, K. J., and D. L. Loftis. 1993. Vegetation diversity after logging in the southern Appalachians. Conservation Biology 7:220–221. Ellison, L. 1960. Influence of grazing on plant succession of rangelands. Botanical Review 26:1–78. Elton, C. S. 1958. The Ecology of Invasions by Animals and Plants. London: Methuen. Elzinga, C. L., D. W. Salzer, and J. W. Willoughby. 1998. Measuring and Monitoring Plant Populations. Report BLM/RS/ST98/005+1730. Denver, CO: U.S. Department of Land Management. ESRI. 1997. ARC/INFO. Redlands, CA: Environmental Systems Research Institute. Evanko, A. B., and R. A. Peterson. 1955. Comparisons of protected and grazed mountain rangelands in southwestern Montana. Ecology 36:71–82. Facelli, J. M., R. J. C. Leon, and V. A. Deregibus. 1989. Community structure in grazed and ungrazed grassland sites in the flooding Pampa, Argentina. American Midland Naturalist 121:125–133. Fairweather, P. G. 1991. Statistical power and design requirements for environmental monitoring. Australian Journal of Marine and Freshwater Research 42:555–567. Fargione, J. E., and D. Tilman. 2005. Diversity decreases invasion via both sampling and complementarity effects. Ecology Letters 8:604–611. Ferreira, L. V. 1997. Effects of the duration of flooding on species richness and floristic composition in three hectares in the Jau National Park in floodplain forests in central Amazonia. Biodiversity and Conservation 6:1353–1363. Ferreira, L. V., and T. J. Stohlgren. 1999. Effects of river level fluctuation in species richness, diversity, and plant distribution in a floodplain in central Amazonia. Oecologia 120:582–587. Fisser, H. G. 1970. Exclosure Studies with Transects of Permanent Plots: 1969 Data. Laramie, WY: Wyoming Agricultural Experiment Station, University of Wyoming.



Fleischner, T. L. 1994. Ecological costs of livestock grazing in western North America. Conservation Biology 8:629–644. Floerl, O., T. K. Pool, and G. J. Inglis. 2004. Positive interactions between nonindigenous species facilitate transport by human vectors. Ecological Applications 14:1724–1736. Foley, J. A. 1994. The sensitivity of the terrestrial biosphere to climate change: a simulation of the middle Holocene. Global Biogeochemical Cycles 8:505–525. Fortin, M. 1994. Edge detection algorithms for two-dimensional ecological data. Ecology 75:956–965. Fortin, M., P. Drapeau, and P. Legendre. 1989. Spatial autocorrelation and sampling design in plant ecology. Vegetatio 83:209–222. Fox, J. F. 1979. Intermediate-disturbance hypothesis. Science 204:1344–1345. Fox, M. D., and B. J. Fox. 1986. The susceptibility of natural communities to invasion. In: R. H. Groves and J. J. Burdon, eds. Ecology of Biological Invasions. Cambridge: Cambridge University Press; pp. 57–60. Francis, R. E., R. S. Driscoll, and J. N. Reppert. 1972. Loop-Frequency as Related to Plant Cover, Herbage Production, and Plant Density. Research Paper RM-94. Fort Collins, CO: Rocky Mountain Forest and Range Experiment Station, U.S. Forest Service. Frank, A. B., D. L. Tanaka, L. Hofmann, and R. F. Follett. 1995. Soil carbon and nitrogen of northern Great Plains grasslands as influenced by longterm grazing. Journal of Range Management 48:470–474. Frank, D. A., and S. J. McNaughton. 1993. Evidence for the promotion of aboveground grassland production by native large herbivores in Yellowstone National Park. Oecologia 96:157–161. Franklin, J. F. 1993. Preserving biodiversity: species, ecosystems, or landscapes? Ecological Applications 3:202–206. Fridley, J. D., R. L. Brown, and J. F. Bruno. 2004. Null models of exotic invasion and scale-dependent patterns of native and exotic species richness. Ecology 85:3215–3222. Fridriksson, S. 1975. Surtsey: Evolution of Life on a Volcanic Island. London: Butterworths. Garcia-Pichel, F., and J. Belnap. 1996. Microenvironments and microscale productivity of cyanobacterial desert crusts. Journal of Phycology 32:774–782. Gaston, K. J. 2000. Global patterns in biodiversity. Nature 405:220–227. Gee, G. W., and J. W. Bauder. 1986. Particle size analysis. In: A. Klute, ed. Methods of Soil Analysis. Part 1—Physical and Mineralogical Methods. Madison, WI: American Society of Agronomy; pp. 383–411. Gentry, A. H. 1988. Tree species richness of upper Amazonian forests. Proceedings of the National Academy of Sciences USA 85:156–159. Gibson, C. W. D., and V. K. Brown. 1991. The effects of grazing on local colonization and extinction during early succession. Journal of Vegetation Science 2:291–300. Giesler, R., M. Högberg, and P. Högberg. 1998. Soil chemistry and plant in Fennoscandian boreal forest as exemplified by a local gradient. Ecology 79:119–137.



Gilbert, B., and M. J. Lechowicz. 2005. Invasibility and abiotic gradients: the positive correlation between native and exotic plant diversity. Ecology 86:1848–1855. Gillison, A. N. 1970. Structural and floristics of a montane grassland/forest transition, Doma Peaks Region, Papua. Blumea 18:71–86. Gleason, H. A. 1917. The structure and development of the plant association. Bulletin of the Torrey Botanical Club 44:463–481. Gleason, H. A. 1920. Some applications of the quadrat method. Bulletin of the Torrey Botanical Club 47:21–33. Gleason, H. A. 1922. On the relation between species and area. Ecology 3:158–162. Gleason, H. A. 1925. Species and area. Ecology 6:66–74. Gleason, H. A. 1926. The individualistic concept of the plant association. Bulletin of the Torrey Botanical Club 53:1–20. Glenn, S. M., and S. L. Collins. 1992. Effects of scale and disturbance on rates of immigration and extinction of species in prairies. Oikos 63:273–280. Glenn-Lewin, D. C. 1977. Species diversity in North American temperate forests. Vegetatio 33:153–162. Goodall, D. W. 1953a. Objective methods for the classification of vegetation, I. The use of positive interspecific correlation. Australian Journal of Botany 1:39–62. Goodall, D. W. 1953b. Objective methods for the classification of vegetation, II. Fidelity and indicator value. Australian Journal of Botany 1:434–456. Goodall, D. W. 1954. Objective methods for the classification of vegetation, III. An essay in the use of factor analysis. Australian Journal of Botany 2:302–324. Goodall, D. W. 1957. Some considerations in the use of point quadrat methods for the analysis of vegetation. Australian Journal of Biological Science 5:1–41. Gosz, J. R. 1991. Fundamental ecological characteristics of landscape boundaries. In: E. Holland, P. G. Risser, and R. J. Naiman, eds. Ecotones: The Role of Landscape Boundaries in the Management and Restoration of Changing Environments. New York: Chapman & Hall; pp. 8–30. Gosz, J. R. 1993. Ecotone hierarchies. Ecological Applications 3:369–376. Graber, D. M., S. A. Haultain, and J. E. Fessenden. 1993. Conducting a biological survey: a case study from Sequoia and Kings Canyon National Parks. In: Proceedings of the Fourth Biennial Conference on Science in California's National Parks. Davis, CA: Cooperative Park Studies Unit, University of California, Davis. Grace, J. B. 1999. The factors controlling species density in herbaceous plant communities. Perspectives in Plant Ecology, Evolution, and Systematics 2:1–28. Grace, J. B., and B. H. Pugesek. 1997. A structural equation model of plant species richness and its application to a coastal wetland. American Naturalist 149:436–460. Green, D. G. 1982. Fire and stability in the postglacial forests of southwest Nova Scotia. Journal of Biogeography 9:29–40.



Green, D. G. 1983. The ecological interpretation of fine resolution pollen records. New Phytologist 94:459–477. Green, D. G. 1989. Simulated effects of fire, dispersal, and spatial pattern on competition within forested mosaics. Vegetatio 82:139–153. Greenberg, C. H., S. H. Crownover, and D. R. Gordon. 1997. Roadside soils: a corridor for invasion of xeric scrub by nonindigenous plants. Natural Areas Journal 17:99–109. Greig-Smith, P. 1957. Quantitative Plant Ecology. London: Butterworths Scientific. Greig-Smith, P. 1964. Quantitative Plant Ecology. London: Butterworths Scientific. Grime, J. P. 1973. Competitive exclusion in herbaceous vegetation. Nature 242:344–347. Grime, J. P. 1977. Evidence for the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist 11:1169–1194. Grime, J. P. 1979. Plant Strategies and Vegetation Processes. New York: John Wiley & Sons. Grubb, P. J. 1977. Maintenance of species-richness in plant communities—importance of regeneration niche. Biological Reviews of the Cambridge Philosophical Society 52:107–145. Guenther, D., T. J. Stohlgren, and P. Evangelista. 2004. A comparison of a near-relict site and a grazed site in a pinyonjuniper community in the Grand Staircase-Escalante National Monument, Utah. In: C. van Riper III and K. L. Cole, eds. The Colorado Plateau: Cultural, Biological, and Physical Research. Tucson, AZ: University of Arizona Press; pp. 153–162. Hadley, K. S., and T. T. Veblen. 1993. Stand response to western spruce budworm and Douglas-fir bark beetle outbreaks, Colorado Front Range. Canadian Journal of Forest Research 23:479–491. Hansen, A. J., and F. di Castri, eds. 1992. Landscape Boundaries: Consequences for Biotic Diversity and Landscape Flows. New York: Springer-Verlag. Hargrove, W. W., and J. Pickering. 1992. Pseudoreplication: a sine qua non for regional ecology. Landscape Ecology 6:251–258. Harper, J. L. 1969. The role of predation in vegetational diversity. Brookhaven Symposium of Biology 22:48–62. Harper, J. L. 1977. Population Biology of Plants. New York: Academic Press. Harrison, S., D. D. Murphy, and P. R. Ehrlich. 1988. Distribution of the bay checkerspot butterfly, Euphydryas editha bayensis: evidence for a metapopulation model. American Naturalist 132:360–382. Hart, R. H., and J. A. Hart. 1997. Rangelands of the Great Plains before European settlement. Rangelands 19:4–11. Haslett, J., and A. E. Raftery. 1989. Space-time modelling with long-memory dependence: assessing Ireland's wind power resource. Applied Statistics 38:1–21. Haury, L. R., J. A. McGowen, and P. H. Wiebe. 1978. Patterns and processes in the time-space scales of plankton distributions. In: J. H. Steele, ed. Spatial Pattern in Plankton Communities. New York: Plenum; pp. 277–327.



Havesi, J. A., J. D. Istok, and A. L. Flint. 1992. Precipitation estimation in mountainous terrain using multivariate geostatistics. Part I: structural analysis. Journal of Applied Meteorology 3:661–676. Hawkins, C. P. 1986. Pseudo-understanding of pseudoreplication: a cautionary note. Bulletin of the Ecological Society of America 67:185. Heady, H. F. 1957. The measurement and value of plant height in the study of herbaceous vegetation. Ecology 38:313–320. Heady, H. F. 1968. Exclosures. In: F. N. Golley and H. K. Buechner, eds. A Practical Guide to the Study of Productivity of Large Herbivores. Oxford: Blackwell Press: pp. 250–252. Heltshe, J. F., and N. E. Forrester. 1983. Estimating species richness using the jackknife procedure. Biometrics 39:1–12. Herben, T., B. Mandák, K. Bímova, and Z. Münzbergova. 2004. Invasibility and species richness of a community: a neutral model and a survey of published data. Ecology 85:3223–3233. Herrick, J. E., W. W. Whitford, A. G. de Soyza, and J. Van Zee. 1996. Soil and vegetation indicators for assessment of rangeland ecological condition. In: North American Workshop on Monitoring for Ecological Assessment of Terrestrial and Aquatic Ecosystems. General Technical Report RM-GTR-284. Fort Collins, CO: Rocky Mountain Forest and Range Experiment Station, U.S. Forest Service. Hickman, J., ed. 1993. The Jepsen Manual: Higher Plants of California. Berkeley: University of California Press. Higgins, S. I., D. M. Richardson, and R. M. Cowling. 1996. Modeling invasive plant spread: the role of plantenvironment interactions and model structure. Ecology 77:2043–2054. Hinds, W. T. 1984. Towards monitoring of long-term trends in terrestrial ecosystems. Environmental Conservation 11:11–18. Hobbs, R. J., and L. F. Huenneke. 1992. Disturbance, diversity, and invasion: implications for conservation. Conservation Biology 6:324–337. Holland, M. M., P. G. Risser, and R. J. Naiman, eds. 1991. Ecotones. New York: Chapman & Hall. Hongo, A., S. Matsumoto, H. Takahashi, H. Zou, J. Cheng, H. Jia, and Z. Zhao. 1995. Effect of exclosure and topography on rehabilitation of overgrazed shrub-steppe in the loess plateau of northwest China. Restoration Ecology 3:18–25. Hormay, A. L. 1949. Getting better records of vegetation changes with the line interception method. Journal of Range Management 2:67–69. Horn, H. S. 1975. Markovian properties of forest succession. In: M. L. Cody and J. M. Diamond, eds. Ecology and Evolution of Communities. Cambridge, MA: Belknap Press; pp. 196–211. Howell, W. 1998. Germination and Establishment of Bromus tectorum L. in Relation to Cation Exchange Capacity, Seedbed, Litter, Soil Cover and Water. Prescott, AZ: Prescott College. Hubbell, S. P. 2001. The Unified Neutral Theory of Biodiversity and Biogeography. Princeton, NJ: Princeton University Press. Hughes, L. E. 1980. Six grazing exclosures with a message. Rangelands 2:17–18.



Hughes, L. E. 1983. Is no grazing really better than grazing? Rangelands 5:159–161. Hughes, L. E. 1990. Twenty years of rest-rotation grazing on the Arizona strip—an observation. Rangelands 12:173–176. Hughes, L. E. 1996. What's in an exclosure? Rangelands 18:201–203. Hulbert, L. C. 1969. Fire and litter effects in undisturbed bluestem prairie in Kansas. Ecology 50:874–877. Hulbert, L. C. 1988. Causes of fire effects in tallgrass prairie. Ecology 46:58. Hulme, P. E. 1996. Herbivores and the performance of grassland plants: a comparison of arthropod, mollusc and rodent herbivory. Journal of Ecology 84:43–51. Hunt, C. B. 1975. Death Valley: Geology, Ecology, and Archeology. Berkeley: University of California Press. Hurlbert, S. H. 1984. Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54:197–211. Huston, M. A. 1979. A general hypothesis of species diversity. American Naturalist 113:81–101. Huston, M. A. 1994. Biological Diversity: The Coexistence of Species in Changing Landscapes. New York: Cambridge University Press. Huston, M. A. 1999. Local processes and regional patterns: appropriate scales for understanding variation in the diversity of plants and animals. Oikos 86:393–401. Huston, M. A., and D. L. DeAngelis. 1994. Competition and coexistence: the effects of resource transport and supply rates. American Naturalist 144: 954–977. Huston, M. A., and A. C. McBride. 2002. Evaluating the relative strengths of biotic versus abiotic controls on ecosystem processes. In: M. Loreau, S. Naeem, and P. Inchausti, eds. Approaches to Understanding Biodiversity and Ecosystem Function. New York: Oxford University Press; pp. 47–60. Hutchings, S. S., and R. C. Holmgren. 1959. Interpretation of loop-frequency data as a measure of plant cover. Ecology 40:668–677. Hutchinson, G. E. 1953. The concept of pattern in ecology. Proceedings of the Philadelphia Academy of Natural Science 105:1–12. Isaaks, E. H., and R. M. Srivastiva. 1980. An Introduction to Applied Geostatistics. New York: Oxford University Press. Jaccard, P. 1901. Etude comparative de la distribution florale dans une portion des Alpes et du Jura. Bulletin Societe Vaudoise des Sciences Naturelles 37:547–579. Jacobson, G. L., T. Webb III, and E. C. Grimm. 1987. Patterns and rates of vegetation change during the glaciation of eastern North America. In: W. F. Ruddiman and E. H. Wright, eds. North America and Adjacent Oceans During the Last Deglaciation. Boulder, CO: Geological Society of America; pp. 277–288. Jensen, J. R. 1996. Introductory Digital Image Processing: A Remote Sensing Perspective, 2nd ed. Upper Saddle River, NJ: Prentice-Hall. Joern, A., and K. H. Keeler, eds. 1995. The Changing Prairie: North American Grasslands. New York: Oxford University Press. Johnston, A. 1957. A comparison of the line interception, vertical point



quadrat, and loop methods as used in measuring basal area of grassland vegetation. Canadian Journal of Plant Science 37:34–42. Jones, J. R., and N. V. DeByle. 1985. Fire. In: Aspen: Ecology and Management in the Western United States. General Technical Report RM-119. Fort Collins, CO: U.S. Forest Service; pp. 77–81. Jones, K. B. 1986. The inventory and monitoring process. In: A. Y. Cooperrider, R. J. Boyde, and H. R. Stuart, eds. Inventory and Monitoring of Wildlife Habitat. Denver, CO: Bureau of Land Management Service Center; pp. 1–10. Junk, W. J. 1989. The use of Amazonian floodplains under an ecological perspective. Interciencia 14:317–322. Kalkhan, M. A., R. M. Reich, and T. J. Stohlgren. 1998. Assessing the accuracy of Landsat Thematic Mapper classification using double sampling. International Journal of Remote Sensing 19:2049–2060. Kalkhan, M. A., and T. J. Stohlgren. 2000. Using multi-scale sampling and spatial cross-correlation to investigate patterns of plant species richness. Environmental Monitoring and Assessment 64:591–605. Kalkhan, M. A., T. J. Stohlgren, and M. B. Coughenour. 1995. An investigation of biodiversity and landscape-scale gap patterns using double sampling: a GIS approach. In: Proceedings of the Ninth Conference on Geographic Information Systems for Natural Resources, Environment, and Land Information Management, Vancouver, British Columbia; pp. 708–712. Kallas, M. 1997. Armillaria root rot disease in the Black Hills National Forest. Master's thesis. Fort Collins: Colorado State University. Kareiva, P. M., and M. Anderson. 1988. Spatial aspects of species interactions: the wedding of models and experiments. In: A. Hastings, ed. Community Ecology. Lecture Notes in Biomathematics 77. Berlin: Springer-Verlag; pp. 35–50. Kaye, M. 2002. Population and age distributions of aspen (Populus tremuloides) in Rocky Mountain National Park, Colorado. PhD dissertation. Fort Collins: Colorado State University. Kaye, M. W., K. Suzuki, D. Binkley, and T. J. Stohlgren. 2001. Landscape-scale dynamics of aspen in Rocky Mountain National Park, Colorado. In: Sustaining Aspen in Western Landscapes: Symposium Proceedings. 13–15 June 2000; Grand Junction, CO. Proceedings RMRS-P-18. Fort Collins, CO: Rocky Mountain Research Station, U.S. Forest Service. Keel, S. H., and G. T. Prance. 1979. Studies of the vegetation of a black water igapó (Rio Negro-Brazil). Acta Amazonia 9:645–655. Keeley, J.E., M. Carrington, and S. Trnka. 1995. Overview of management issues raised by the 1993 wildfires in southern California. In: J. E. Keeley and T. Scott, eds. Brushfires in California: Ecology and Resource Management. Fairfield, WA: International Association of Wildland Fire; pp. 83–89. Keeley, J. E., D. Lubin, and C. J. Fotheringham. 2003. Fire and grazing impacts on plant diversity and alien plant invasions in the southern Sierra Nevada. Ecological Applications 13:1355–1374. Keeley, J. E., and Fotheringham, C. J. 2005. Plot shape effects on plant species diversity measurements. Journal of Vegetation Science 16:249–256.



Kempton, R. A. 1979. Structure of species abundance and measurement of diversity. Biometrics 35:307–321. Kerner von Marilaun, A. 1863/1951. Plant Life of the Danube Basin. Translated by H. S. Conrad. Ames: Iowa State College Press. Kershaw, K. A. 1963. Patterns in vegetation and its causality. Ecology 44:377–388. Kieft, T. L., C. S. White, S. R. Loftin, R. Aguilar, J. A. Craig, and D. A. Skaar. 1998. Temporal dynamics in soil carbon and nitrogen resources at a grass-land-shrubland ecotone. Ecology 79:671–683. Kikkawa, J., and E. E. Williams. 1971. Altitudinal distribution of land birds in New Guinea. Search 2:64–69. Kindschy, R. R. 1987. Sagehen exclosure: a history of bitterbrush reproduction. Rangelands 9:113–114. Knight, D. H. 1994. Mountains and Plains: The Ecology of Wyoming Landscapes. New Haven, CT: Yale University Press. Kot, M., M. A. Lewis, and P. van den Driessche. 1996. Dispersal data and the spread of invading organisms. Ecology 77:2027–2042. Kozlowski, T. T. 1982. Water supply and tree growth. Part II. Flooding. Forest Abstracts 43:145–161. Kozlowski, T. T. 1984. Responses of woody plants to flooding. In: T. T. Kozlowski, ed. Flooding and Plant Growth. New York: Academic Press; pp. 129–164. Krebs, C. J. 1989. Ecological Methodology. New York: Harper & Row. Krebs, C. J. 1999 Ecological Methodology. Menlo Park, CA: Addison-Wesley Educational Publishers. Kullman, L. 1996. Rise and demise of cold-climate Picea abies forest in Sweden. New Phytologist 134:243–256. Kupper, L. L., and K. B. Hafner. 1989. How appropriate are popular sample size formulas? American Statistician 43:101–105. Lacey, C. A. 1989. Knapweed management: a decade of change. In: P. K. Kay and J. R. Lacey, eds. Proceedings of Knapweed Symposium, April 4–5, 1989. Extension Bulletin 45, Plant and Soil Science Department and Cooperative Extension Service. Bozeman: Montana State University; pp. 1–6. Lacey, C. A., P. Husby, and G. Handl. 1990. Observations on spotted and diffuse knapweed into ungrazed bunchgrass communities in western Montana. Rangelands 12:30–32. LaRoe, E. T. 1993. Implementation of an ecosystem approach to endangered species conservation. Endangered Species Update 10:3–6. Larsen, K. D. 1995. Effects of microbiotic crusts on the germination and establishment of three range grasses. Unpublished thesis. Boise, ID: Boise State University. Lavado, R. S., J. O. Sierra, and P. N. Hashimoto. 1996. Impact of grazing on soil nutrients in a Pampean grassland. Journal of Range Management 49:452–457. Lavoie, C., M. Jean, F. Delisle, and G. Létourneau. 2003. Exotic plant species of the St. Lawrence River wetlands: a spatial and historical analysis. Journal of Biogeography 30:537–549. Law, R., and R. D. Morton. 1996. Permanence and the assembly of ecological communities. Ecology 77:762–775.



Le, N. D., and J. V. Zidek. 1992. Interpolation with uncertain spatial covariances: a Bayesian alternative to kriging. Journal of Multivariate Analysis 43:351–374. Leach, M. K., and T. J. Givnish. 1996. Ecological determinants of species loss in remnant prairies. Science 273:1555–1558. Lee, M. 2001. Non-native plant invasions in Rocky Mountain National Park: linking species traits and habitat characteristics. Master's thesis. Fort Collins: Colorado State University. Legendre, P., and M. J. Fortin. 1989. Spatial pattern and ecological analysis. Vegetatio 80:107–138. Levin, S. A. 1992. The problem of pattern and scale in ecology. Ecology 73:1943–1967. Levine, J. M. 2000. Species diversity and biological invasions: relating local process to community pattern. Science 288:852–854. Levine, J. M., and C. M. D'Antonio. 1999. Elton revisited: a review of evidence linking diversity and invasibility. Oikos 87:15–26. Li, J., A. Herlihy, W. Gerth, P. Kaufmann, S. Gregory, S. Urquhart, and D. P. Larsen. 2001. Variability in stream macroinvertebrates at multiple spatial scales. Freshwater Biology 46:87–97. Likens, G. E., ed. 1989. Long-Term Studies in Ecology: Approaches and Alternatives. New York: Springer-Verlag. Lister, A. J., P. P. Mou, R. H. Jones, and R. J. Mitchell. 2000. Spatial patterns of soil and vegetation in a 40-year-old slash pine (Pinus elliottii) forest in the Coastal Plain of South Carolina, U.S.A. Canadian Journal of Forest Research 30: 145–155. Lonsdale, W. M. 1999. Global patterns of plant invasions and the concept of invasibility. Ecology 80:1522–1536. Loope, L. L., and D. Mueller-Dombois. 1989. Characteristics of invaded islands, with special reference to Hawaii. In: J. A. Drake, F. DiCasti, R. H. Groves, F. J. Kruger, H. A. Mooney, M. Rejmánek, and M. H. Williams, eds. Biological Invasions: A Global Perspective. New York: John Wiley & Sons; pp. 257–280. Lubchenco, J., A. M. Olson, L. B. Brubaker, S. R. Carpenter, M. M. Holland, S. P. Hubbell, S. A. Levin, J. A. MacMahon, P. A. Matson, J. M. Melillo, H. A. Mooney, C. H. Peterson, H. R. Pulliam, L. A. Real, P. J. Regal, and P. G. Risser. 1991. The sustainable biosphere initiative: an ecological research agenda. Ecology 72:371–412. Ludwig, J. A., and J. F. Reynolds. 1988. Statistical Ecology: A Primer on Methods and Computing. New York: John Wiley & Sons. MacArthur, R. H., and E. O. Wilson. 1967. The Theory of Island Biogeography. Princeton, NJ: Princeton University Press. MacDonald, L. H., A. W. Smart, and R. C. Wissmar. 1991. Monitoring guidelines to evaluate effects of forestry activities on streams in the Pacific Northwest and Alaska. EPA/910/9–91–001. Seattle: U.S. Environmental Protection Agency. Mack, R. N. 1981. Invasion of Bromus tectorum L. into western North America: an ecological chronicle. AgroEcosystems 7:145–165. Mack, R. N. 1996. Understanding the processes of weed invasions: the influence of environmental stochasticity. In: C. H. Stirton, ed. Weeds in a



Changing World. Symposium Proceedings no. 64. Brighton, England: British Crop Protection Council; pp. 65–74. Mack, R. N., D. Simberloff, M. Lonsdale, H. Evans, M. Clout, and F. Bazzaz. 2000. Biotic invasions: causes, epidemiology, global consequences and control. Issues in Ecology 5:1–20. Mack, R. N., and J. N. Thompson. 1982. Evolution in steppe with few large, hooved mammals. American Naturalist 119:757–773. Madole, R. F. 1976. Glacial geology of the Front Range, Colorado. In: W. C. Mahaney, ed. Quaternary Stratigraphy of North America. Stroudsburg, PA: Dowden, Hutchinson, and Ross; pp. 319–351. Magnuson, J. J., T. K. Kratz, T. M. Frost, C. J. Bowser, B. J. Benson, and R. Nero. 1991. Expanding the temporal and spatial scales of ecological research and comparison of divergent ecosystems: roles for LTER in the United States. In: P. G. Risser, ed. Long-Term Ecological Research: An International Perspective. New York: John Wiley & Sons; pp. 45–70. Magurran, A. E. 1988. Ecological Diversity and Its Measurement. Princeton, NJ: Princeton University Press. Magurran, A. E. 2003. Measuring Biological Diversity. Malden, MA: Blackwell. Malanson, G. P. 1993. Riparian Landscapes. Cambridge: Cambridge University Press. Margules, C. R., and R. L. Pressey. 2000. Systematic conservation planning. Nature 405:243–253. Markgraf, V., and L. Scott. 1981. Lower timberline in central Colorado during the past 15,000 yr. Geology 9:231–234. Martinez, E., and E. Fuentes. 1993. Can we extrapolate the California model of grassland-shrubland ecotone? Ecological Applications 3:417–423. MathSoft Inc. 1999. S-Plus 2000. Seattle: MathSoft, Inc. Matlack, G. R. 1994. Vegetation dynamics of the forest edgetrends in space and successional time. Journal of Ecology 82:113–123. Matson, P. A., and S. R. Carpenter. 1990. Special feature: Statistical analysis of ecological response to large-scale perturbations. Ecology 71:2037–2068. May, R. M. 1973. Stability and Complexity in Model Ecosystems. Princeton, NJ: Princeton University Press. May, R. M. 1999. Unanswered questions in ecology. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 354: 1951–1959. McCune, B. 1997. Influence of noisy environmental data on canonical correspondence analysis. Ecology 78:2617–2623. McGowan, J. A. 1990. Climate and change in oceanic systems: the value of time series data. Trends in Ecology and Evolution 5:293–299. McIntosh, R. P. 1985. The Background of Ecology: Concepts and Theory. New York: Cambridge University Press. McNaughton, S. J. 1979. Grassland-herbivore dynamics. In: A. R. E. Sinclair and M. Norton-Griffiths, eds. Serengeti: Dynamics of an Ecosystem. Chicago: University of Chicago Press; pp. 46–81. McNaughton, S. J. 1983. Serengeti grassland ecology: the role of composite environmental factors and contingency in community organization. Ecological Monographs 53:291–320.



McNaughton, S. J. 1993. Biodiversity and function of grazing ecosystems. In: E. D. Schulze and H. A. Mooney, eds. Biodiversity and Ecosystem Function. Berlin: Springer-Verlag; pp. 361–408. McNaughton, S. J., M. Oesterheld, D. A. Frank, and K. J. Williams. 1989. Ecosystem-level patterns of primary productivity and herbivory in terrestrial habitats. Nature 132:142–144. Messer, J. J., R. A. Linthurst, and W. S. Overton. 1991. An EPA program for monitoring ecological status and trends. Environmental Monitoring and Assessment 17:67–78. Metzger, K. 1997. Modeling forest stand structure to a ten meter resolution using Landsat TM data. Master's thesis. Fort Collins: Colorado State University. Meyers, N., R. A. Mittermeier, C. G. Mittermeier, G. A. B. da Fonseca, and J. Kent. 2000. Biodiversity hotspots for conservation priorities. Nature 403:853–858. Milchunas, D. G., and W. K. Lauenroth. 1993. Quantitative effects of grazing on vegetation and soils over a global range of environments. Ecological Monographs 63:327–366. Milchunas, D. G., W. K. Lauenroth, P. L. Chapman, and M. K. Kazempour. 1990. Community attributes along a perturbation gradient in a shortgrass steppe. Journal of Vegetation Science 1:375–384. Milchunas, D. G., O. E. Sala, and W. K. Lauenroth. 1988. A generalized model of the effects of grazing by large herbivores on grassland community structure. American Naturalist 132:87–106. Miller, R. I., and R. G. Wiegert. 1989. Documenting completeness, species-area relations, and the species-abundance distribution of a regional flora. Ecology 70:16–22. Milne, B. T., A. R. Johnson, T. H. Keith, C. A. Hatfield, J. David, and P. T. Hraber. 1996. Detection of critical densities associated with piñon-juniper woodland ecotones. Ecology 77:805–821. Mistry, S., J. Comiskey, and T. J. Stohlgren. 1999. In: A. Alonso and F. Dallmeier, eds. Biodiversity Assessment and Monitoring of the Urumbaba Region, Peru. SI/MAB Series 3, Smithsonian Institution Monitoring and Assessment of Biodiversity Program. Washington, DC: Smithsonian Institution. Monk, C. D. 1967. Tree species diversity in the eastern deciduous forest with particular reference to north central Florida. American Naturalist 101: 173–187. Montaña, C., J. Lopez-Portillo, and A. Mauchamp. 1990. The response of two woody species to the conditions created by a shifting ecotone in an arid ecosystem. Journal of Ecology 78:789–798. Moser, E. B., A. M. Saxton, and S. R. Pezeshki. 1990. Repeated measures analysis of variance: application to tree research. Canadian Journal of Forest Research 20:524–535. Mueggler, W. F. 1985. Forage. In: Aspen: Ecology and Management in the Western United States. General Technical Report RM-119. Fort Collins, CO: U.S. Forest Service; pp. 129–134. Mueller-Dombois, D., and H. Ellenberg. 1974. Aims and Methods of Vegetation Ecology. New York: John Wiley & Sons.



Mulder, B. S., B. R. Noon, T. A. Spies, M. G. Raphael, C. J. Palmer, A. R. Olsen, G. H. Reeves, and H. H. Welsh. 1999. The strategy and design for the effectiveness monitoring program for the Northwest Forest Plan. General Technical Report PNW-GTR-437. Portland, OR: U.S. Forest Service. Munz, P. A., and D. D. Keck. 1959. A California Flora and Supplement. Berkeley: University of California Press. Myers, N., R. A. Mittermeier, C. G. Mittermeier, G. A. B. da Fonseca, and J. Kent. 2000. Biodiversity hotspots for conservation priorities. Nature 403: 853–858. Nasri, M., and P. S. Doescher. 1995. Effect of competition by cheatgrass on shoot growth of Idaho fescue. Journal of Range Management 48:402–405. National Research Council. 1990. Forest Research: A Mandate for Change. Washington, DC: National Academies Press. National Research Council. 1994. Rangeland Health: New Methods to Classify, Inventory, and Monitor Rangelands. Washington, DC: National Academies Press. Neilson, R. P. 1991. Climatic constraints and issues of scale controlling regional biomes. In: M. M. Holland, R. J. Naiman, and P. G. Risser, eds. Role of Landscape Boundaries in the Management and Restoration of Changing Environments. New York: Chapman & Hall; pp. 31–51. Neilson, R. P. 1995. A model for predicting continental-scale vegetation distribution and water balance. Ecological Applications 5:362–385. Nero, R. W., and J. J. Magnuson. 1992. Effects of changing spatial scale on acoustic observations of patchiness in the Gulf Stream. Landscape Ecology 6:279–291. Neter, J., W. Wasserman, and M. H. Kutner. 1990. Applied Linear Statistical Models: Regression, Analysis of Variance, and Experimental Designs, 3rd ed. Homewood, IL: Irwin. Nevah, Z., and R. H. Whittaker. 1979. Structural and floristic diversity of shrublands and woodlands in northern Israel and other Mediterranean areas. Vegetatio 41:171–190. Nichols, H. 1982. Review of late Quaternary history of vegetation and climate in the mountains of Colorado. In: J. C. Halfpenny, ed. Ecological Studies in the Colorado Alpine: A Festschrift for John W. Marr. Occasional Paper 37. Boulder: Institute of Arctic and Alpine Research, University of Colorado; pp. 27–33. Nichols, J. D. 1992. Capture-recapture models: using marked animals to study population dynamics. BioScience 42:94–102. Niklas, K. J., B. H. Tiffney, and A. H. Knoll. 1983. Patterns in vascular land plant diversification. Nature 303:293–299. Noss, R. 1983. A regional landscape approach to maintain diversity. BioScience 33:700–706. Noss, R. F., and A. Y. Cooperrider. 1994. Saving Nature's Legacy: Protecting and Restoring Biodiversity. Washington, DC: Island Press. Nusser, S. M., and J. J. Goebel. 1997. The National Resources Inventory: a long-term multi-resource monitoring programme. Environmental and Ecological Systematics 4:181–204. Nusser S. M., F. J. Breidt, and W. A. Fuller. 1998. Design and estimation for



investigating the dynamics of natural resources. Ecological Applications 8:234–245. Oosting, H. G. 1948. The Study of Plant Communities. San Francisco: Freeman. Opdam, P., R. Van Apeldoorn, A. Schotman, and J. Kalkhoven. 1993. Population responses to landscape fragmentation. In: C. C. Vos and P. Opdam, eds. Landscape Ecology of a Stressed Environment. New York: Chapman & Hall; pp. 147–171. Orr, D. M., and C. J. Evenson. 1991. Effects of sheep grazing Astrebla grasslands in central western Queensland III. Dynamics of Astrebla spp. under grazing and exclosure between 1975 and 1986. Rangeland Journal 13:36–46. Pacala, S. W., and M. J. Crawley. 1992. Herbivores and plant diversity. American Naturalist 140:243–260. Palmer, C. J., K. H. Ritters, J. Strickland, D. C. Cassell, G. E. Byers, M. L. Papp, and C. I. Liff. 1991. Monitoring and Research Strategy for Forests—Environmental Monitoring and Assessment Program (EMAP). EPA/600/4–91/012. Washington, DC: U.S. Environmental Protection Agency. Palmer, M. W. 1990. The estimation of species richness by extrapolation. Ecology 71:1195–1198. Palmer, M. W. 1993. Putting things in even better order: the advantages of canonical correspondence analysis. Ecology 74:2215–2230. Palmer, M. W. 1994. Variation in species richness: toward a unification of hypotheses. Folia Geobotany Phytotax Praha 29:511–530. Palmer, M. W., and P. M. Dixon. 1990. Small-scale environmental heterogeneity and the analysis of species distributions along gradients. Journal of Vegetation Science 1:57–65. Palmer, M. W., and G. Rusch. 2001. How fast is the carousel? Direct indices of species mobility in an Oklahoma grassland. Journal of Vegetation Science 12:305–318. Parker, K. W. 1951. A Method for Measuring Trends in Range Condition in National Forest Ranges. Washington, DC: U.S. Forest Service. Parsons, A. J., A. Harvey, and I. R. Johnson. 1991. Plant-animal interactions in a continuously grazed mixture. II. The role of differences in the physiology of plant growth and of selective grazing on the performance and stability of species in a mixture. Journal of Applied Ecology 28:635–647. Peet, R. K. 1981. Forest vegetation of the Colorado Front Range: patterns of species diversity. Vegetatio 52:129–140. Peet, R. K. 1988. Forests of the Rocky Mountains. In: M. G. Barbour and W. D. Billings, eds. North American Terrestrial Vegetation. New York: Cambridge University Press; pp. 64–103. Penfound, W. T. 1963. A modification of the point-centered quadrat method for grassland analysis. Ecology 44:175–176. Petchey, O. L., A. Hector, and K. J. Gaston. 2004. How do different measures of functional diversity perform? Ecology 85:847–857. Peters, A., D. E. Johnson, and M. R. George. 1996. Barb goatgrass: a threat to California rangelands. Journal of Range Management 18:8–10.



Peters, R. L., and J. D. Darling. 1985. The greenhouse effect and nature reserves. BioScience 35:707–717. Peters, R. L., and T. E. Lovejoy, eds. 1992. Global Warming and Biological Diversity. New Haven, CT: Yale University Press. Peterson, D. L., and V. T. Parker, eds. 1998. Ecological Scale: Theory and Applications. New York: Columbia University Press. Petterson, E. S. 1999. Prescribed fire effects on plant communities in Rocky Mountain bighorn sheep habitat. Master's thesis. Fort Collins: Colorado State University. Pianka, E. 1980. Guild structure in desert lizards. Oikos 35:194–201. Pickett, S. T. A. 1989. Space-for-time substitution as an alternative to long-term studies. In: G. E. Likens, ed. Long-Term Studies in Ecology: Approaches and Alternatives. New York: Springer-Verlag; pp. 110–135. Pickett, S. T. A., and P. S. White, eds. 1985. The Ecology of Natural Disturbance and Patch Dynamics. New York: Academic Press. Pielou, E. C. 1969. An Introduction to Mathematical Ecology. New York: John Wiley & Sons. Pielou, E. C. 1977. Mathematical Ecology. New York: John Wiley & Sons. Pielou, E. C. 1991. After the Ice Age: The Return of Life to Glaciated North America. Chicago: University of Chicago Press. Pitelka, L. F., R. H. Gardner, J. Ash, S. Berry, H. Gitay, I. R. Noble, A. Saunders, R. H. W. Bradshaw, L. Brubaker, J. S. Clark, M. B. Davis, S. Sugita, J. M. Dyer, R. Hengeveld, G. Hope, B. Huntley, G. A. King, S. Lavorel, R. N. Mack, G. P. Malanson, M. McGlone, I. C. Prentice, and M. Rejmanek. 1997. Plant migration and climate change. American Scientist 85:464–473. Planty-Tabacchi, A.-M., E. Tabacchi, R. J. Naiman, C. DeFerrari, and H. Decamps. 1996. Invasibility of species-rich communities in riparian zones. Conservation Biology 10:598–607. Podani, J., T. Czaran, and S. Bartha. 1993. Pattern, area and diversity: the importance of spatial scale in species assemblages. Abstracta Botanica 17: 35–51. Pole, A., M. West, and P. J. Harrison. 1994. Applied Bayesian Forecasting and Time Series Analysis. New York: Chapman & Hall. Polley, H. W., and S. L. Collins. 1984. Relationships of vegetation and environment in buffalo wallows. American Midland Naturalist 112:178–186. Polley, H. W., and L. L. Wallace. 1986. The relationship of plant species heterogeneity to soil in buffalo wallows. Southwestern Naturalist 31:493–501. Pollock, M. M., R. J. Naiman, and T. A. Hanley. 1998. Plant species richness in riparian wetlands: a test of biodiversity theory. Ecology 79:94–105. Post, W. M., and S. L. Pimm. 1983. Community variability and food-web stability. Mathematical Biosciences 64:169–192. Pound, R., and F. E. Clements. 1898a. A method of determining the abundance of secondary species. Minnesota Botanical Gardens 2:19–24. Pound, R., and F. E. Clements. 1898b. A Phytogeography of Nebraska. New York: Arno Press. Powell, T. M. 1989. Physical and biological scales of variability in lakes, estuaries, and the coastal ocean. In: J. Roughgarden, R. M. May, and S. A.



Levin, eds. Perspectives in Ecological Theory. Princeton, NJ: Princeton University Press; pp. 157–176. Powell, T. M., and J. H. Steele, eds. 1994. Ecological Time Series. New York: Chapman & Hall. Preston, F. W. 1960. Time and space and the variation of species. Ecology 41:611–627. Preston, F. W. 1962a. Canonical distribution of commonness and rarity: part I. Ecology 43:185–215. Preston, F. W. 1962b. Canonical distribution of commonness and rarity: part II. Ecology 43:410–432. Preston, F. W. 1969. Diversity and stability in the biological world. Brook-haven Symposia in Biology 22:1–12. Pugesek, B. H., A. Tomer, and A. Von Eye, eds. 2003. Structural Equation Modeling: Applications in Ecological and Evolutionary Biology. New York: Cambridge University Press. Pulliam, H. R. 1988. Sources, sinks, and population regulation. American Naturalist 132:652–661. Puyravaud, J., J. Pascal, and C. Dufour. 1994. Ecotone structure as an indicator of changing forest-savanna boundaries (Linganamakki Region, southern India). Journal of Biogeography 21:581–593. Ramensky, L. G. 1924. Die Grundgesetzmassigkeiten im Aufbau der Vegetationsdecke. Botan. Centralbl. N. F. 7:453–455. Randall, J. M. 1996. Weed control and the preservation of biological diversity. Weed Technology 10:370–383. Raunkiaer, C. 1934. The Life Forms of Plants and Statistical Plant Geography: Being the Collected Papers of C. Raunkiaer. Oxford: Clarendon Press. Reardon, J. J. 1996. Changes in grazed and protected plant communities in Yellowstone National Park. In: F. J. Singer, ed. Effects of Grazing by Wild Ungulates in Yellowstone National Park. Washington, DC: U.S. Department of the Interior, National Park Service, Natural Resource Program Center, Natural Resource Information Division,; pp. 115–125 Reed, R. A., R. K. Peet, M. W. Palmer, and P. S. White. 1993. Scale dependence of vegetation-environment correlations: a case study of a North Carolina piedmont woodland. Journal of Vegetation Science 4:329–340. Reich, R. M., and V. A. Bravo. 1998. Integrating spatial statistics with GIS and remote sensing in designing multiresource inventories. Presented at the North American Symposium on Toward a Unified Framework for Inventorying and Monitoring Forest Ecosystem Resources, Guadalajara, Mexico, November 1–6, 1998. Reich, R. M., R. L. Czaplewski, and W. A. Bechtold. 1994. Spatial cross-correlation in growth of undisturbed natural shortleaf pine stands in northern Georgia. Journal of Environmental and Ecological Statistics 1:201–217. Reichard, S. H., and C. W. Hamilton. 1997. Predicting invasions of woody plants introduced into North America. Conservation Biology 11:193–203. Reichard, S. H., and P. White. 2001. Horticulture as a pathway of invasive plant introductions in the United States. BioScience 51:103–113. Reid, W. V., and K. R. Miller. 1989. Keeping Options Alive: The Scientific Basis for Conserving Biodiversity. Washington, DC: World Resources Institute.



Rejmánek, M. 1996. Species richness and resistance to invasions. In: H. Orians, R. Dirzo, and J. H. Cushman, eds. Biodiversity and Ecosystem Processes in Tropical Forests. Berlin: Springer-Verlag; pp. 153–172. Reynolds, R. T., and C. H. Trost. 1980. The response of native vertebrate populations to crested wheatgrass planting and grazing by sheep. Journal of Range Management 33:122–125. Rice, B., and M. Westoby. 1983. Plant species richness at the 0.1 hectare scale in Australian vegetation compared to other continents. Vegetatio 52:129–140. Rice, E. L. 1952. Phytosociologic analysis of a tall-grass prairie in Marshall County, Oklahoma. Ecology 33:112–115. Riegel, G. M., S. E. Green, M. E. Harmon, and J. F. Franklin. 1988. Characteristics of mixed conifer forest reference stands at Sequoia National Park, California. Technical Report no. 32. Davis, CA: Cooperative National Park Resources Studies Unit, University of California, Davis. Riitters, K. H., B. E. Law, R. C. Kucera, A. L. Gallant, R. L. DeVelice, and C. J. Palmer. 1992. A selection of forest condition indicators for monitoring. Environmental Monitoring and Assessment 20:21–23. Risser, P. G. 1993. Ecotones: ecotones at local to regional scales from around the world. Ecological Applications 3:367–368. Robertson, G. P. 1987. Geostatistics in ecology: interpolating with known variance. Ecology 68:744–748. Robertson, J. H. 1971. Changes on a sagebrush-grass range in Nevada ungrazed for 30 years. Journal of Range Management 24:397–400. Robinson, G. R., and J. F. Quinn. 1988. Extinction, turnover and species diversity in an experimentally fragmented California annual grassland. Oecologia 76:71–82. Robinson, G. R., J. F. Quinn, and M. L. Stanton. 1995. Invasibility of experimental habitat in California winter annual grassland. Ecology 76:786–794. Romme, W. H., and M. G. Turner. 1991. Implications of global climate change for biogeographic patterns in the Greater Yellowstone ecosystem. Conservation Biology 5:373–386. Romme, W. H., M. G. Turner, L. L. Wallace, and J. S. Walker. 1995. Aspen, elk, and fire in northern Yellowstone National Park. Ecology 76:2097–2106. Rosentreter, R. 1994. Displacement of rare plants by exotic grasses. In: S. B. Monsen and S. G. Kitchen, eds. Proceedings—Ecology and Management of Annual Rangelands. General Technical Report INT-GTR-313. Ogden, UT: Intermountain Research Station, U.S. Forest Service; pp. 170–175. Rosenzweig, M. L. 1995. Species Diversity in Space and Time. Cambridge: Cambridge University Press. Rosenzweig, M. L., and Y. Ziv. 1999. The echo pattern of species diversity: pattern and processes. Ecography 22:614–628. Rummel, D. J., and J. Roughgarden. 1983. Some differences between invasion-structured and co-evolutionarystructured competitive communities. Oikos 41:477–486. Rummell, R. S. 1951. Some effects of grazing on ponderosa pine forest and range in central Washington. Ecology 32:594–607. Rusek, J. 1993. Air-pollution-mediated changes in alpine ecosystems and ecotones. Ecological Applications 3:406–416.



Salo, J., R. Kalliola, Y. Makinem, P. Niemela, M. Puhakka, and P. D. Coleu. 1986. River dynamics and the diversity of Amazon lowland forest. Nature 322:254–258. Salt, G. W. 1957. An analysis of avifaunas in the Teton Mountains and Jackson Hole, Wyoming. Condor 59:373–393. Saltonstall, K. 2002. Cryptic invasion by a non-native genotype of the common reed, Phragmites australis, into North America. Proceedings of the National Academy of Science USA 99:2445–2449. SAS Institute. 1998. SAS for Windows. Cary, NC: SAS Institute. Sax, D. F., and S. D. Gaines. 2003. Species diversity: from global decreases to local increases. Trends in Ecology and Evolution 18:561–566. Schreuder, H. T., M. S. Williams, and R. M. Reich. 1999. Estimating the number of tree species in a forest community using survey data. Environmental Monitoring and Assessment 56:293–303. Schulz, T. T., and W. C. Leininger. 1990. Differences in riparian vegetation structure between grazing areas and exclosures. Journal of Range Management 43:295–299. Scott, C. T. 1998. Sampling methods for estimating change in forest resources. Ecological Applications 8:228–233. Scott, J. M., F. Davis, R. Csuti, R. Noss, B. Butterfield, C. Groves, H. Anderson, S. Caicco, F. D'Erchia, T. C. Edwards, Jr., J. Ulliman, and R. G. Wright. 1993. GAP analysis: a geographic approach to protection of biological diversity. Wildlife Monographs 123:1–41. Seastedt, T. R. 1995. Soil systems and nutrient cycles of the North American prairie. In: A. Joern and K. H. Keeler, eds. The Changing Prairie: North American Grasslands. New York: Oxford University Press; pp. 157–174. Shafer, C. L. 1990. Nature Reserves: Island Theory and Conservation Practice. Washington, DC: Smithsonian Institution Press Sheley, R. L., B. E. Olson, and L. L. Larson. 1997. Effect of weed seed rate and grass defoliation on diffuse knapweed. Journal of Range Management 50:39–43. Shmida, A. 1984. Whittaker's plant diversity sampling method. Israel Journal of Botany 33:41–46. Shmida, A., and M. V. Wilson. 1985. Biological determinants of species diversity. Journal of Biogeography 12:1–20. Short, H. L., and J. B. Hestbeck. 1995. National biotic resource inventories and GAP analysis: problems of scale and unproven assumptions limit a national program. BioScience 45:535–539. Shultz, L. M. 1998. The flora and fauna of the Colorado Plateau: what do we know? In: L. M. Hill, ed. Learning from the Land, Grand Staircase-Escalante National Monument Science Symposium. Salt Lake City, UT: Paragon Press; pp. 203–210. Simonson, S. 1998. Rapid assessment of butterfly diversity: a method for landscape assessment. Master's thesis. Fort Collins, CO: Colorado State University. Simonson, S., G. Chong, P. Opler, and T. J. Stohlgren. 2001. Rapid assessment of butterfly diversity: a method for landscape evaluation. Biodiversity and Conservation 10:1369–1386. Singer, F. J. 1995. Effects of grazing by ungulates on upland bunchgrass



communities of the northern winter range of Yellowstone National Park. Northwest Science 69:191–203. Singer, F. J. 1996. Differences between willow communities browsed by elk and communities protected for 32 years in Yellowstone National Park. In: F. J. Singer, ed. Effects of Grazing by Wild Ungulates in Yellowstone National Park. Washington, DC: U.S. Department of the Interior, National Park Service, Natural Resource Program Center, Natural Resource Information Division; pp. 279–290. Smeins, F. E., T. W. Taylor, and L. B. Merrill. 1976. Vegetation of a 25-year exclosure on the Edwards Plateau, Texas. Journal of Range Management 29:24–29. Smith, D. A., and E. M. Schmutz. 1975. Vegetative changes on protected versus grazed desert grassland ranges in Arizona. Journal of Range Management 28:453–458. Smith, D. R. 1960. Description and response to elk use of two mesic grassland and shrub communities in the Jackson Hole region of Wyoming. Northwest Science 34:25–36. Smith, M. D., J. C. Wilcox, T. Kelly, and A. K. Knapp. 2004. Dominance not richness determines invasibility of tallgrass prairie. Oikos 106:253–262. Sneva, F. A., L. R. Rittenhouse, P. T. Tueller, and P. Reece. 1984. Changes in protected and grazed sagebrush-grass in eastern Oregon, 1937 to 1974. Station Bulletin 663. Corvallis: Agricultural Experiment Station, Oregon State University; pp. 3–11. Sokal, R. R., and F. J. Rohlf. 1981. Biometry: The Principles and Practice of Statistics on Biological Research, 2nd ed. San Francisco: W. H. Freeman. Sokal, R. R., and J. D. Thomson. 1987. Applications of spatial autocorrelation in ecology. In: P. Legendre and L. Legendre, eds. Developments in Numerical Taxonomy. NATO ASI Series, vol. 14. Berlin: Springer-Verlag; pp. 431–466. Soulé, M. E., and K. A. Kohm. 1989. Research Priorities for Conservation Biology. Washington, DC: Island Press. SPSS Inc. 1997. Systat, version 7.0. Chicago: SPSS Inc. SPSS Inc. 2001. Systat, version 10.0. Chicago: SPSS Inc. Stapanian, M. A., S. P. Cline, and D. C. Cassell. 1993. Vegetation structure. In: B. L. Conkling and G. E. Byers, eds. Forest Health Monitoring Field Methods Guide. Internal Report. Las Vegas: U.S. Environmental Protection Agency. Stebler, F. G., and C. Schröter. 1892. Versuch einer übersicht über die Wiesentypen der Schweiz. Landwirtschaftliches Jahrbuch der Schweiz 6:95. Steele, J. H. 1989. Discussion: scale and coupling in ecological systems. In: J. Roughgarden, R. M. May, and S. A. Levin, eds. Perspectives in Ecological Theory. Princeton, NJ: Princeton University Press; pp. 177–180. Stohlgren, T. J. 1992. Resilience of a heavily logged grove of giant sequoia (Sequoiadendron giganteum) in Sequoia and Kings Canyon National Parks. Forest Ecology and Management 54:115–140. Stohlgren, T. J. 1993. Bald eagle winter roost characteristics in Lava Beds National Monument, California. Northwest Science 67:44–54.



Stohlgren, T. J. 1994. Planning long-term vegetation studies at landscape scales. In: T. M. Powell and J. H. Steele, eds. Ecological Time Series. New York: Chapman & Hall; pp. 208–241. Stohlgren, T. J. 1999a. Measuring and monitoring biodiversity in forests and grasslands in the United States. In: C. Aguirre and C. R. Franco, eds. Proceedings of the North American Symposium Toward a Unified Framework for Inventory and Monitoring Forest Ecosystem Resources. General Technical Report RMRS-P-12. Fort Collins, CO: Rocky Mountain Forest and Range Experiment Station, U.S. Forest Service; pp. 248–255. Stohlgren, T. J. 1999b. The Rocky Mountains. In: M. J. Mac, P. A. Opler, C. E. Puckett Haecher, and P. D. Doran, eds. Status and Trends of the Nation's Biological Resources. Reston, VA: Biological Resources Division, U.S. Geological Survey; pp. 473–504. Stohlgren, T. J. 2001. Data acquisition for ecological assessments. In: M. Jensen and P. Bougeron, eds. Ecological Assessments. New York: Springer-Verlag; pp. 71–78. Stohlgren, T. J. 2002. Beyond theories of plant invasions: lessons from natural landscapes. Comments on Theoretical Biology 7:355–379. Stohlgren, T. J., and R. R. Bachand. 1997. Lodgepole pine (Pinus contorta) ecotones in Rocky Mountain National Park, Colorado, USA. Ecology 78:632–641. Stohlgren, T. J., R. R. Bachand, Y. Onami, and D. Binkley. 1998a. Species-environment relationships and vegetation patterns: effects of spatial scale and tree life-stage. Plant Ecology 135:215–228. Stohlgren, T. J., D. T. Barnett, C. Flather, J. Kartesz, and B. Peterjohn. 2005a. Plant species invasions along the latitudinal gradient in the United States. Ecology 86:2298–2309. Stohlgren, T. J., J. Belnap, G. W. Chong, and R. Reich. 1998b. A plan to assess native and exotic plant diversity and cryptobiotic crusts in the Grand Staircase-Escalante National Monument. In: L. M. Hill, ed. Learning from the Land, Grand Staircase-Escalante National Monument Science Symposium Proceedings. Salt Lake City, UT: Paragon Press; pp. 269–276. Stohlgren, T. J., D. Binkley, G. W. Chong, M. A. Kalkhan, L. D. Schell, K. A. Bull, Y. Otsuki, G. Newman, M. Bashkin, and Y. Son. 1999a. Exotic plant species invade hot spots of native plant diversity. Ecological Monographs 69:25–46. Stohlgren, T. J., D. Binkley, T. T. Veblen, and W. L. Baker. 1995a. Attributes of landscape-scale, long-term studies: malpractice insurance for landscape ecologists. Environmental Monitoring and Assessment 36:1–25. Stohlgren, T. J., K. A. Bull, and Y. Otsuki. 1998c. Comparison of rangeland vegetation sampling techniques in the Central Grasslands. Journal of Range Management 51:164–172. Stohlgren, T. J., K. A. Bull, Y. Otsuki, C. A. Villa, and M. Lee. 1998d. Riparian zones as havens for exotic plant species. Plant Ecology 138:113–125. Stohlgren, T. J., T. N. Chase, R. A. Pielke, Sr., T. G. F. Kittel, and J. S. Baron. 1998e. Evidence that local land use practices influence regional climate, vegetation, and stream flow patterns in adjacent natural areas. Global Change Biology 4:495–504.



Stohlgren, T. J., G. W. Chong, M. A. Kalkhan, and L. D. Schell. 1997a. Multiscale sampling of plant diversity: effects of minimum mapping unit size. Ecological Applications 7:1064–1074. Stohlgren, T. J., G. W. Chong, M. A. Kalkhan, and L. D. Schell. 1997b. Rapid assessment of plant diversity patterns: a methodology for landscapes. Environmental Monitoring and Assessment 48:25–43. Stohlgren, T. J., G. W. Chong, L. D. Schell, K. A. Rimar, Y. Otsuki, M. Lee, M. A. Kalkhan, and C. A. Villa. 2002. Assessing vulnerability to invasion by nonnative plant species at multiple spatial scales. Environmental Management 29:566–577. Stohlgren, T. J., M. B. Coughenour, G. W. Chong, D. Binkley, M. A. Kalkhan, L. D. Schell, D. J. Buckley, and J. K. Berry. 1997c. Landscape analysis of plant diversity. Landscape Ecology 12:155–170. Stohlgren, T. J., C. Crosier, G. Chong, D. Guenther, and P. Evangelista. 2005c. Life-history habitat matching in invading non-native plant species. Plant and Soil 277:7–18. Stohlgren, T. J., M. B. Falkner, and L. D. Schell. 1995b. A modified-Whittaker nested vegetation sampling method. Vegetatio 117:113–121. Stohlgren, T. J., D. Guenther, P. Evangelista, and N. Alley. 2005b. Patterns of plant rarity, endemism, and uniqueness in an arid landscape. Ecological Applications 15:715–725. Stohlgren, T. J., M. W. Kaye, A. D. McCrumb, Y. Otsuki, B. Pfister, and C. A. Villa. 2000a. Using new video mapping technology in landscape ecology. BioScience 50:529–536. Stohlgren, T. J., Y. Otsuki, C. A. Villa, M. Lee, and J. Belnap. 2001. Patterns of plant invasions: a case example in native species hotspots and rare habitats. Biological Invasions 3:37–50. Stohlgren, T. J., A. J. Owen, and M. Lee. 2000b. Monitoring shifts in plant diversity in response to climate change: a method for landscapes. Bio-diversity and Conservation 9:65–86. Stohlgren, T. J., and J. F. Quinn. 1992. An assessment of biotic inventories in western U.S. national parks. Natural Areas Journal 12:145–154. Stohlgren, T. J., J. F. Quinn, M. Ruggiero, and G. S. Waggoner. 1995c. Status of biotic inventories in US National Parks. Biological Conservation 71:97–106. Stohlgren, T. J., L. D. Schell, and B. Vanden Heuvel. 1999b. How grazing and soil quality affect native and exotic plant diversity in Rocky Mountain grasslands. Ecological Applications 9:45–64. Stolte, K. W. 1997. 1996 National Technical Report on Forest Health. Report FS-605. Asheville, NC: Southern Research Station, U.S. Forest Service. Stoms, D. M. 1992. Effects of habitat map generalization in biodiversity assessment. Photogrammetric Engineering and Remote Sensing 58:1587–1591. Strayer, D., J. S. Glitzenstein, C. G. Jones, J. Kolasa, G. E. Lichens, M. J. McDonnell, G. G. Parker, and S. T. A. Pickett. 1986. Long-Term Ecological Studies: An Illustrated Account of Their Design, Operation, and Importance to Ecology. Occasional Publication no. 2. Millbrook, NY: Institute of Ecosystem Studies.



Stubbendieck, J., and G. D. Willson. 1987. Prairie resources of national park units in the Great Plains. Natural Areas Journal 7:100–106. Sukachev, V. 1945. Biogeocoenology and phytocoenology. Comptes Rendus Academie des Sciences de U.S.S.R. 47:429–431. Sukachev, V., and N. Dylis. 1964. Fundamentals of Forest Biogeocoenology. Edinburgh: Oliver & Boyd. Suzuki, K. 1997. Aspen regeneration in elk winter range of Rocky Mountain National Park and Roosevelt National Forest, Colorado. Master's thesis. Fort Collins: Colorado State University. Suzuki, K., H. Suzuki, D. Binkley, and T. J. Stohlgren. 1999. Aspen regeneration in the Colorado Front Range: differences at local and landscape scales. Landscape Ecology 14:231–237. Swengel, A. B., and S. R. Swengel. 1995. The tallgrass prairie butterfly community. In: Our Living Resources: A Report to the Nation on the Distribution, Abundance, and Health of U.S. Plants, Animals, and Ecosystems. Washington, DC: National Biological Service, U.S. Department of the Interior; pp. 174–176. Sykes, M. T., and I. C. Prentice. 1996. Climate change, tree species distributions and forest dynamics: a case study in the mixed conifer northern hardwoods zone of northern Europe. Climatic Change 34:161–177. Takeuchi, M. 1962. The structure of the Amazonian vegetation. VI. Igapó. Journal Of the Faculty of Science University of Tokyo III 8:297–301. Tansley, A. G., and T. F. Chipp, eds. 1926. Aims and Methods in the Study of Vegetation. London: British Empire Vegetation Committee and Crown Agents for the Colonies. ter Braak, C. J. F. 1986. Canonical correspondence analysis: a new eigen vector technique for multivariate direct gradient analysis. Ecology 67: 1167–1179. ter Braak, C. J. F. 1987a. The analysis of vegetation-environment relationships by canonical correspondence analysis. Vegetatio 69:69–77. ter Braak, C. J. F. 1987b. CANOCO—a FORTRAN program for canonical community ordination by [partial] [detrended] [canonical] correspondence analysis, principal components analysis and redundancy analysis. Wageningen, The Netherlands: TNO Institute of Applied Computer Science. ter Braak, C. J. F. 1991. CANOCO. Wageningen, The Netherlands: Agricultural Mathematics Group. Thomas, L., and C. J. Krebs. 1997. A review of statistical power analysis software. Bulletin of the Ecological Society of America 78:128–139. Tiedemann, A. R., and H. W. Berndt. 1972. Vegetation and soils of a 30-year deer and elk exclosure in central Washington. Northwest Science 46:59–66. Tilman, D. 1982. Resource Competition and Community Structure. Monographs in Population Biology. Princeton, NJ: Princeton University Press. Tilman, D. 1988. Plant Strategies and the Structure and Dynamics of Plant Communities. Princeton, NJ: Princeton University Press. Tilman, D. 1997. Community invasibility, recruitment limitation, and grassland biodiversity. Ecology 78:81–92. Tilman, D. 1999. The ecological consequences of changes in biodiversity: a search for general principles. Ecology 80:1455–1475.



Tilman, D., W. Wedin, and J. Knops. 1996. Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature 379:718–720. Timmins, S. M., and P. A. Williams. 1991. Weed numbers in New Zealand's forest and scrub reserves. New Zealand Journal of Ecology 15:153–162. Tobey, R. 1981. Saving the Prairies: The Life Cycle of the Founding School of American Plant Ecology, 1985–1955. Berkeley: University of California Press. Turelli, M. 1981. Niche overlap and invasion of competitors in random environments. Models without demographic stochasticity. Theoretical Population Biology 20:1–56. Turner, M. B. 1989. Landscape ecology: the effect of pattern on process. Annual Review of Ecological Systematics 20:171–197. Twery, M. J., G. A. Elmes, and C. B. Yuill. 1991. Scientific exploration with an intelligent GIS: predicting species composition from topography. AI Applications 5:45–53. U.S. Forest Service. 1985. Range Analysis and Management Handbook. FSH 2209.21. Amendment 15. Washington, DC: U. S. Forest Service. U.S. Forest Service. 1996. Range Analysis and Management Training Guide. Lakewood, CO: Rocky Mountain Region, U.S. Forest Service. U.S. General Accounting Office. 1991. Rangeland management: comparison of rangeland condition reports. GAO/ RCED-91–191. Washington, DC: U.S. General Accounting Office. U.S. Soil Conservation Service. 1976. National Range Handbook, Washington, DC: U.S. Soil Conservation Service. van der Maarel, E., and M. T. Sykes. 1993. Small-scale plant species turnover in a limestone grassland: the carousel model and some comments on the niche concept. Journal of Vegetation Science 4:179–188. Vestal, A. G. 1943. Unequal scales for rating species in communities. American Journal of Botany 30:305–310. Vitousek, P. M. 1990. Biological invasions and ecosystem processes: towards an integration of population biology and ecosystem studies. Oikos 57:7–13. Walker, J., and R. K. Peet. 1984. Composition and species diversity of pinewiregrass savannas of the Green Swamp, North Carolina. Vegetatio 55: 163–179. Walter, H. 1964. Die Vegetation der Erde in öko-physiologischer Betrachtung. Band I: Die tropischen und subtropischen Zonen, 2nd ed. Jena, Germany: VEB Fischer. Walter, H. 1971. Ecology of Tropical and Subtropical Vegetation. Translated by D. Mueller-Dombois, edited by J. H. Burnett. Edinburgh: Oliver & Boyd. Weaver, J. E., and F. E. Clements. 1938. Plant Ecology, 2nd ed. New York: McGraw-Hill. Weaver, J. E., and W. W. Hansen. 1941. Native Midwestern Pastures—Their Origin, Composition, and Degeneration. Bulletin 22. Lincoln: University of Nebraska Conservation and Survey Division. Webster, D. B. 1992. Viewpoint: replication, randomization, and statistics in range science. Journal of Range Management 45:285–290. Weisberg, P. J., and W. L. Baker. 1995. Spatial variation in tree seedlings



and krummholz growth in the forest-tundra ecotone of Rocky Mountain National Park, Colorado, U.S.A. Arctic and Alpine Research 27:116–129. Weitz, A., D. Bunte, and H. Hersemann. 1993. Application of nested sampling technique to determine the scale of variation in soil physical and chemical properties. Catena 20: 207–714. Welsh, S. L., and N. D. Atwood. 1998. Flora of Bureau of Land Management Grand Staircase-Escalante National Monument. Kanab, UT: Bureau of Land Management, Kanab. Wesser, S. D., and W. S. Armbruster. 1991. Species distribution controls across a forest-steppe transition: a causal model and experimental test. Ecological Monographs 61:323–342. West, M., and P. J. Harrison. 1997. Bayesian Forecasting and Dynamic Models. New York: Springer-Verlag. West, N. E., K. H. Rea, and R. O. Harniss. 1979. Plant demographic studies in sagebrush-grass communities of southeastern Idaho. Ecology 60:376–388. Westbrooks, R. 1998. Invasive Plants, Changing the Landscape of America: Fact Book. Washington, DC: Federal Interagency Committee for the Management of Noxious and Exotic Weeds. Whicker, A. D., and J. K. Detling. 1988. Ecological consequences of prairie dog disturbances. BioScience 38:778–785. Whisenant, S. G., and D. W. Uresk. 1990. Spring burning Japanese brome in a western wheatgrass community. Journal of Range Management 43:205–208. White, R. S., and P. O. Currie. 1983. Prescribed burning in the northern Great Plains: yield and cover responses of 3 forage species in the mixed grass prairie. Journal of Range Management 36:179–183. Whittaker, R. H. 1962. Classification of natural communities. Botanical Review 28:1–239. Whittaker, R. H. 1965. Dominance and diversity in land plant communities. Science 147:250–260. Whittaker, R. H. 1967. Gradient analysis of vegetation. Biological Review 42:207–264. Whittaker, R. H. 1970. Communities and Ecosystems. London: Macmillan. Whittaker, R. H. 1977. Evolution of species diversity on land communities. Evolutionary Biology 10:1–67. Whittaker, R. H., and W. A. Niering. 1975. Vegetation of Santa Catalina Mountains, Arizona. V. Biomass, production, and diversity along the elevation gradient. Ecology 56:771–790. Whittaker, R. H., W. A. Niering, and M. O. Crisp. 1979. Structure, pattern, and diversity of a mallee community in New South Wales. Vegetatio 39:65–76. Wiens, J. A., C. S. Crawford, and J. R. Gosz. 1985. Boundary dynamics: a conceptual framework for studying landscape ecosystems. Oikos 45:421–427. Willems, J. H. 1980. Observations on northwest European limestone grassland communities. 5.A. An experimental approach to the study of species-diversity and above-ground biomass in chalk grassland. Proceedings of the Koninklijke Nederlandse Akademie Van Wetenschappen Series C-Biological and Medical Sciences 83:279–295.



Wilson, E. O. 1988. Biodiversity. Washington, DC: National Academies Press. Winegar, H. H. 1977. Camp Creek channel fencing—plant, wildlife, soil, and water responses. Rangeman's Journal 4:10–12. Wiser, S. K., R. K. Peet, and P. S. White. 1998. Prediction of rare-plant occurrence: a southern Appalachian example. Ecological Applications 8: 909–920. Woods, K. D., and M. B. Davis. 1989. Paleoecology of range limits—beech in the Upper Peninsula of Michigan. Ecology 70:681–696. Woodward, A., E. G. Schreiner, D. B. Houston, and B. B. Moorhead. 1994. Ungulate-forest relationships in Olympic National Park: retrospective exclosure studies. Northwest Science 68:97–110. Woodward, F. I. 1993. The lowland-to-upland transition—modeling plant responses to environmental change. Ecological Applications 3:404–408. Worster, D. 1977. Nature's Economy: A History of Ecological Ideas. New York: Cambridge University Press. Young, V. A. 1943. Changes in vegetation and soil of palouse prairie caused by overgrazing. Journal of Forestry 41:834–838. Zar, J. H. 1974. Biostatistical Analysis. Englewood Cliffs, NJ: Prentice-Hall. Zar, J. H. 1996. Biostatistical Analysis, 3rd edition. Upper Saddle River, NJ: Prentice-Hall. Zedaker, S. M., and N. S. Nicholas. 1990. Quality Assurance Methods Manual for Forest Site Classification and Field Measurements. EPA/600/3–90/082. Corvallis, OR: U.S. Environmental Protection Agency.

Index abiotic factors. See environmental factors abundance, index of, 18 accessibility sampling, 194 accuracy, 57, 80, 256 aerial photographs, interpretation of, 144–145 agricultural fields, chronosequence studies in abandoned, 280 Agropyron spicatum, 206, 216 Aims and Methods of Vegetation Ecology, 36 air pollution, 274 Akaike Information Criteria, 311 allelopathy, 28 alpha diversity, 43, 289 alpha level, 60, 62 altitude. See elevation ancillary data, 277 anisotropy, 261 arid environments, 218; phenology, 6 aridity, species richness and, 227, 236 aspen regeneration, plots for evaluating, 319 aspen vegetation type, 300; invasibility, 233 associations, 16; classification of, 30; Clements ideas on, 18; Daubenmire's focus on, 25; reality of, 38 Bailey, Bob, 18 Barbour, M. G., 43 bare ground; percentage of, 203, 226; species richness and, 227, 230 Bell and Lechowicz plots, 107, 109 belt transects, 22, 76 beta diversity, 43, 289 beta level, 60, 62 bias, 278 biodiversity; effect of invasive species on, 171; effect of scale and resolution, 145; sustaining,



121; underestimating, 133; unified theory of, 44 biogeography, unified theory of, 44 biomass; above-ground, 121, 187; effect of grazing on, 195; peaks, 208; sampling, 30, 55 biomes, 18; testing multiple, 115–116 Biota of North America Program, 163 blooming, multiple, 6 botanists, on field crews, 160–161 boundaries, 67, 256, 289; of vegetation types, 152 Braun-Blanquet, Josias, 18, 21 Bray, J. R., 24 Bromus tectorum, 207, 212, 215, 235; spread of, 287 canonical correspondence analysis, 222, 245–248, 282, 291; of ecotones, 294–296 canopy cover, 27 carbon, in soil, 208; species richness and, 227–228, 246 Carrington plots, 109 cause-and-effect relationships, 231, 241 Centaurea diffusa, 212 cheatgrass. See Bromus tectorum Chipp, T. F., 22 chronic issues strategy, 274 chronosequence studies, 59, 79, 275–276; in agricultural fields, 280 clans, 22 classification; life form, 20; preoccupation with, 285; system of von Humboldt, 16 clay, species richness and, 177, 182, 227 Clements, Frederick Edward, 18–20; philosophy of sampling, 35 climate; effect on dominant species, 12; effect on species richness, 182, 214–215, 298, 300; effect on tree line, 281; factors, 176, 196 climate change, 274; ability to adapt to, 303; detecting migrations in response to, 300–301; indicators of, 290; monitoring responses to, 287 climax communities, 22, 31; mapping potential, 33 climax stand, 25 cluster analysis, 248 cluster sampling, 279 cokriging, 261, 266–267 colonization rates, 211 communities; classification and emphasis on, 85; Clements' view of, 35; climax, 22, 31; comparisons of richness, 112; contribution to total diversity, 145; Daubenmire's definition of, 25; formation of, 24; open and closed, 18; rare plant, 36; reality of, 38; stability of, 17, 188; Whittaker's view of, 22–23 community succession theory, 188 competition, 18; direct, 9; effect of grazing on, 195; microenvironments, 18; of native versus exotic species, 172, 190; relation to native species richness, 186;

Warming's observations, 17–18 competitive exclusion, 188, 215; effect of grazing on, 209; in exclosures, 195, 200; time needed for, 210–211

Index complementary species, 9 complementary variables, 200 completeness, assessments of, 277 confidence limits, 60, 62 conservation, ecosystem approach, 219 consociations, 22 constancy, 29 contagious distribution, 28 Cooper, W. S., 23 correlations, 239–242 corridors, 156–157 cost efficiency; effect of spatial autocorrelation on, 121; of plant diversity studies, 91; of sampling techniques, 134–137 cover. See also foliar cover; forb cover; canopy, 27; classes, 76, 163; estimates, 115, 163; exotic species and, 163, 234 Cowles, Henry Chandler, 20 cross-correlation procedures, 240 cryptobiotic soil crusts, 218; invasibility and, 235; role in soil stabilization, 219; species richness and, 225–228, 244; vegetation type and, 225 curlycup gumweed. See Gindelia squarrosa current and emerging issues strategy, 273 Curtis, J. T., 24, 28 Darwin, Charles, 16 data. See also data collection; data transformation; comparability, 112; completeness and accuracy, 256; population, 252–253; products in study design, 48; summarizing across sites, 127–128, 130–131 databases, species additions from, 277 data collection; Braun-Blanquet's methods, 21; in field, 258–259; frequency of, 68–69 data transformation, 198, 240 Daubenmire, Rexford, 24–35; philosophy of sampling, 35 Daubenmire transects, 118–120; capture of plant diversity, 128; versus Parker transects, 122–125 Davenport, C. B., 23 day length, 43 density, estimates of, 26 desert ecosystems, 218 detectability, assumption of equal, 285 detection; of exotic species, 127, 132, 163–164, 316; quadrat size and, 160; of rare species, 73, 103, 125, 133–134, 156, 313, 317–318; with various plot designs, 313–314 diffuse knapweed. See Centaurea diffusa digital elevation model, 258 distribution, 20–21, 108. See also patchy distribution; concept of, 284; contagious, 28; effect of environmental changes on, 300–301; evaluating with exclosure studies, 198–199; evaluating with nested plots, 319; on grazed versus ungrazed sites, 212–214; modeling, 277, 321;


quantifying, 251; random versus nonrandom, 21; shifts in, 301; of tree species, 290–291 disturbances, 11, 274, 284; effect on species richness, 43, 186–187, 215, 300; in grasslands,



131; immigration and, 214; invasive species and, 176, 234; plant distribution and, 213; resilience to, 219; transect techniques and, 130 diversity. See biodiversity; plant diversity; species richness diversity indexes, 19, 81, 199, 202–203 diversity-stability hypothesis, 68 dominance. See also dominant species; consequences of emphasis on, 38–39; describing, 98; foliar cover and, 27; sampling techniques and, 131–133 dominance-density curves, 98 dominance-diversity curves, 79 dominant species, 12, 248; assessing cover of, 118; effect of elevation on, 246; effect of environmental factors, 294–296; effect on diversity, 79; overemphasis on, 285–286; retrospective studies of, 274 double sampling, 80, 307–308, 319 Drude, Oscar, 18 ease of analysis, in single-scale plot studies, 90 ease of use, of sampling technique, 136 ecological plasticity, 85 ecoregional maps, 18 ecotones, 22; Daubenmire's recognition of, 25; distribution and, 281; importance of monitoring, 303; models of structure, 302; species composition overlap between, 293; studies of, 288–289 edge effects, 193 education, taxonomy, 5 elevation; concept of zones, 18; effect on invasive species, 187; effect on species richness, 42, 205–206, 242, 264, 298; effect on understory species distribution, 294–295; exotic species richness and, 182, 183, 229, 243–244, 263, 301–302 Ellenberg, Heinz, 36–38 emigration, 8 endemism, 43 environmental factors; across sites, 255; effect on species richness, 298, 300; effect on vegetation patterns, 246; relation to biotic factors, 245; relation to understory composition, 292, 294–295; role in invasibility, 186–187; species classification according to response to, 85; tree distributions and, 290–291 environmental gradients; addressing, 302; assumptions about, 63; avoiding, 288; plant distribution and, 213; plant diversity and, 108; recognition of, 18 environmental stress, species richness and, 42 Eragrostis lehmanniana, 212 error, sources of, 56 error terms, 259 Estimate-S, 65 “-etum,” 17 evenness index, 199 evolution, 43; lag times, 12; primary units, 42

exclosure studies, 191; drawbacks of, location, 194, 196–197

193–196; plot

Index exotic species. See also exotic species richness; competition and, 190; correlation with rare species, 188; cover and, 163, 234; detection methods, 127, 132, 316; early detection of, 163–164; effect of grazing on, 167, 211–212, 216; monitoring, 137, 301–302 exotic species richness; across state plots, 164; analysis of, 249; disturbances and, 186–187; effect of grazing on, 201–202; factors associated with, 182–185, 243, 263; quantifying patterns of, 228–230; relation to native species richness, 165–168, 170, 172; soil characteristics and, 177, 246; variance in, 185; in washes, 225 experimental approach, xii, 117 expertise, lack of taxonomy, 5 extent, 49–50 extinction, 8; rates, 211 extirpation, 8, 170, 212 extrapolation; ability of sampling technique, 134; difficulty from long-term plots, 278; as part of monitoring studies, 277 fertilization, plant diversity and, 43 Festuca idahoensis, 216 fidelity, 30 field crews; expertise, 73; of Forest Health Monitoring Program, 160 field data, acquisition aspects, 258–259 fine-scale variables, 267 fire, 273; monitoring effects, 280; spread of invasive species and, 302 flooding, 81, 234; duration, 85; effect on species richness, 215; tolerance to, 83 flora, local versus regional, 5 foliar cover, 176; effect of grazing on, 216; effect of sampling techniques, 126; estimates of dominance from, 27; in exclosure studies, 201, 203; on grazed versus ungrazed sites, 213; as measure of abundance, 199; relation to species richness, 166, 184; species turnover and, 234 forage production, 121, 193 forb cover, on grazed versus ungrazed sites, 213 forb frequency, effect of grazing on, 203–204 Forest Health Monitoring Program, 87, 106–109, 159–160, 278–279; methods and study areas, 161–163 Forest Inventory and Analysis Program, 107, 159 forward stepwise regression, 176, 243–244 fragile habitats, 52 frames, 136 calibrating, 162 funding; design comparison studies and, 116; scale and, 50 gap-phase dynamics, 23 genotypes, spatial distributions, 303




geographic barriers, to migration, 12 geographic information systems, 38, 134 geographic modeling, 276 geography, 17, 42, 182, 315; oversimplification of, 39–40 geostatistical analysis, 257–261 germination, 8; soil crust cover and, 235 Gindelia squarrosa, 206 Gleason, Henry, 18, 20–21, 42, 101, 304; sampling philosophy, 35 Global Positioning System, use to establish plots, 108 Goodall, D. W., 24 gradient analysis, 222, 245; Whittaker's approach, 23 grain size, 97 Grand Staircase-Esclante National Monument, 220 grazing, 273; assessment of effects, 118, 191–192; directional responses to, 215–217; effect on biomass, 195; effect on exotic species, 167, 186, 211–212; effect on invasibility, 187; effect on soil quality, 199–200; effect on species composition overlap, 204–205; effect on species richness, 201, 209–211; regimes, 194; species, 196; speciesspecific response to, 206–207 Greig-Smith, P., 24 ground truth plots, 147, 148 habitats; characteristics and invasibility, 187; diversity, 68; evaluating differences and similarities, 82, 84–85; fragile, 52; rare, 150; types, 24–25 Harper, John, 42 Hawaiian silver sword alliance, 5 herbivory, 6. See also grazing; variability in, 213 heterogeneity, 131, 154, 248; Daubenmire's observations, 30–31; detected with nested plots, 317; in exclosure studies, 194; gradients, 42; microsite, 79; reducing, 34; relation to native species richness, 186; species composition overlap and, 85; understanding, 284–285 Hill's ratio, 199, 202 homogeneity, 29; assumption of, 40; emphasis on, 120 hot spots; analysis of, 219; detecting, 319; of invasion, 170 Hubbell, Stephen P., 44–45 hydrophytes, 17 identification. See plant identification; taxonomy immigration, 8, 214; balance with extirpation, 212; rates, 211 impact studies, 59 importance values, 28, 98 independent variables; identifying, 249; linear combination of, 259 index of abundance, 18 individual-based reaction-diffusion, 267 individualist concept, 18 initial conditions, assumptions about, 275 insect outbreaks, 215

Index invasibility; assessment of, 250; degree of, 233; effect of grazing on, 212; effect of soil phosphorus on, 228; established exotic species and, 184; factors affecting studies of, 174; factors determining, 186–187, 234–235; role of species turnover, 188, 190; species richness and, 172–173, 185–186 invasive species, 159, 171, 250, 273. See also exotic species; detecting, 103, 132, 163–164; disturbance and, 176; effect of grazing on, 195–196, 201–202; factors correlated with success of, 177–185; monitoring, 167–168, 170, 301–302; patchy distribution of, 165; resistance to, 172 inverse distance sampling, 260 isoclines, 256 Jaccard, P., 101 Jaccard's coefficient, 103, 106, 148, 150, 198, 291 judgmental sampling, 278 Keeley design, 105, 107, 109 Kentucky bluegrass. See Poa pratensis Krebs, C. J., 43 kriging, 89, 222, 256–267; example results, 264–266 K strategies, 18 lag effects, 43 Land Condition Trend Analysis, 80, 278–279 land managers, accommodating needs of, 116 Landsat bands, 262 Landsat classification map, 307 Landsat Thematic Mapper, 258 land use, 273; evaluation of patterns, 121; relation to native species richness, 186 latitude; diversity and, 42; exotic species richness and, 182 Lehmann lovegrass. See Eragrostis lehmanniana life forms; classification, 20; diversification of, 31; effect of grazing on composition, 212–214, 216 life spans, 9 linear regression, 176, 182 linear relationships, 239 linear transects, 76 line intercept technique, 27, 279 local diversity, 27 logistic regression, 250–252 longitude; exotic species richness and, 182; variation in species-area curves and, 315 long-term studies, 12, 14 magnesium, in soil, native species richness and, 228–229, 231 Magurran, A. E., 43 Man and the Biosphere Biological Diversity Program, 277–278 maps; assessment of accuracy, 33; needed improvements of, 155; resolution and species richness, 153; types of, 254–255; use to evaluate plant diversity, 156–158 McIntosh, R. P., 24, 28

Merriam, Clinton Hart, 18 mesophytes, 17 meta-analysis, 59




methodology; calibration, 277; standardized, 111 microhabitats, species composition overlap and, 150 migration; detecting, 300–301; lag times, 12; monitoring, 287–288; relation to native species richness, 186 minimum detectable change, 60–62 minimum mapping units, 144–145, 146; evaluating effects of, 148; number of polygons and, 156; size of, 147; vegetation patterns dependent on, 152 models, predictive capability, 269–270 monitoring, 276–277; assessment of invasive species and, 167–168; baseline data for, 298, 300; expense of, 5–6; national-scale, 170; strategies for, 273–274, 279; of tree species distribution, 280–283; using line intercept methods, 27 montane vegetation, concept of elevation zones, 18 Monte Carlo techniques, 65, 222, 245–246, 270 Moran's I, 310, 313 morphology, link to physiology, 17 Mueller-Dombois, Dieter, 36–38 multiphase design, 270, 276 multiple regression analysis, 242–243, 312 multiscale plots, 220. See also Whittaker plots; advantages of, 317 multiscale sampling, 92, 104–108, 148; assessing, 165; benefits of, 109; effectiveness of, 154–156; in exclosure studies, 195; for exotic species, 137; testing, 143–145 multivariate analysis, 243 native species, protecting, 121, 167 native species richness; effect of grazing on, 201–202, 209–211, 216; in exclosures, 195, 200; exotic species richness and, 165–168, 170, 172, 182–183; predictors of, 296–297; quantifying patterns of, 228–230; regression analysis of, 262–264; relation to invasibility, 185–186; soil characteristics and, 177 nested intensity designs, 276; applications of, 321–322 nested plots, 32, 92–99, 310. See also Whittaker plots; advantages and disadvantages of, 316–318; benefits of, 318–320; circular, 93; example of, 102–104; sampling time, 313; single scale, 86–88; species accumulation curves and, 320–321 niches, 24; differentiation, 188; preemption hypothesis, 79 nitrogen, 231; effect on invasive species, 187; effect on species richness, 177, 182–183, 206, 215, 227, 235, 240–241; in grazed sites, 208 nonindependent points, 93, 114–115 nonmetric multidimensional scaling, 83 nonnative species. See exotic species

Index nonspatial statistical modeling, 239 normal distribution, 198, 240; assumption of, 63 observational studies, 34 observer bias, 277 Oosting, H. G., 23 order in nature, 17 oversampling, 32, 66 overstory species, changes in distributions, 301 pack rat middens, 275 Paczoski, Jozef, 18 paired-plot designs, 191, 208 Parker transects, 118; adoption of, 120; capture of plant diversity, 128; versus Daubenmire transects, 122–125 patchy distribution, 8, 12, 154, 214; detecting species with, 103, 125; in grasslands, 131; of invasive species, 165 path coefficient analysis, 243–244 pathogens, 45 peak phenology, 6 Penfound, W. T., 28 perimeter:area ratio, 96; plot shape and, 112 periodicity, 21 peripheral areas, 288 permanent plots, monitoring, 276–277 phase transition theory, 302 phenology, difficulties in, 5–7 phenotypic variation, 5 phosphorus, species richness and, 227–230, 240–241, 243, 246 photosynthetically active radiation, 242, 246; effect on species richness, 298, 301; effect on understory species distribution, 294–295; index of, 291–292 physiology, recognition of link to morphology, 17 plant diversity. See also species richness; baseline data for monitoring, 298, 300; chronosequence studies of, 275–276; comparisons of vascular, 191; evaluation in exclosure studies, 198–199; importance of estimating, 141; long-term changes, 12, 14; measuring, xi, 163–164; monitoring, 167; patterns among states, 164–165; use of high-resolution maps to evaluate, 156–158; vegetation structure and, 160 plant diversity studies; comparisons of, 209; constraints of long-term, 283–284; design of, xii, 8–14, 116; difficulty of, 4–8; objectives of, 3, 47–48, 252; scope, 48; site selection, 220; temporal, 277–278 plant ecology, history of, 15–24 plant identification, 4, 163, 197–198, 322. See also taxonomy; in Forest Health Monitoring Program, 162 plant mortality, causes of, 9 PLANTS database, 162 plots. See also plot size; assumptions of use, 278; design for species inventories, 309–310; distance effects, 151, 152, 205; establishing, 282; monitoring permanent, 276–277;

383 number needed, 307, 321; paired, 191, 208; placement of, 36, 56, 66, 68, 135, 196–197, 248; randomization, 106, 108; relocatable,



113; selection of, 156; shape, 51–54, 112, 114, 120; subdivision of, 105 plot size, 75–76, 154, 248, 308; consistent proportions, 113; detection of exotic species and, 316; in exclosure studies, 194; inadequacy of, 288; life form diversity and, 31; shape interactions, 114; species composition overlap and, 150; species richness and, 26, 101, 320–321 Poa pratensis, 207, 212 point distribution maps, 254, 255 point frequency sampling, 76 point methods, 26, 28 pollen stratigraphy, 274 polyclaves, 5 ponderosa pine vegetation type, invasibility, 234 population data, 252–253 potential vegetation maps, 18 Pound, Roscoe, 18 precipitation; effect on dominant species, 12; effect on invasive species, 187; effect on species richness, 182, 215, 225 precision, 57, 76, 80 prediction; by models, 269–270; of rare plant distribution, 18 productivity, diversity and, 42 propagule pressure, 186 proportion reduced error values, 250 pseudoreplication, 58–59; in exclosure studies, 193–194 quadrats; Clement's use of, 18–19; Daubenmire's use of, 24; early use of standardized, 18; Gleason's use of, 21; number in single-scale samples, 75–76; placement of, 55; Raunkaier's use of, 20; size effects, 28, 97, 100, 160, 164; use with belt transects, 22 Ramensky, Leonid, 18 randomization, of subplot placement, 116 random sampling, 37–38 Range Analysis and Management Handbook, 120 rangeland; conservation, 121; Daubenmire's impact on sampling, 24; management, 133; native species cover, 166; sampling, 118–120 rare habitats, 216, 217 rare species, 36, 157–158; correlation with exotic species, 188; detecting, 73, 103, 125, 133–134, 156, 313, 317–318; monitoring, 285; points needed to capture, 26; surveying, 7–8; use of spatial patterns to protect, 168 Raunkaier, Christen, 20 reference stands, 87 refuges, 213 regression analysis, 312; assessing spatial autocorrelation, 261; example results, 262–264; explanation of variance by, 108; limitations of, 231–236; linear, 176, 182; logistic, 250–252; multiple, 242–243, 312; simple, 239–242 regression coefficients, 243, 259

regression tree analysis, 248–250 relevé method, 21, 75

Index remote sensing data, 38, 143–144, 267; use to select sample sites, 147 replacement, 9, 31, 188, 190 replication, 58–59, 115, 304; in exclosure studies, 193; requirements for adequate, 116 residuals, of regression models, 259 resolution, 49–50, 269; of invasibility studies, 174; species richness and, 153; testing importance of, 145–147; vegetation types and, 152 resource partitioning, 188 resources; importance for invasive species, 236; maps of, 269; spatial coverage, 258; timing of use, 25 retrospective studies, 274 riparian vegetation type, invasibility, 234 river levels, 81 Rosenzweig, Michael, 44 r strategies, 18 ruderals, 18 running means, 63–64 sample size, 24; in exclosure studies, 193; formula-based determinations, 56, 58, 60; selecting, 54–56; in singlescale plots, 89; species richness and, 43 sampling. See also sampling design; sampling techniques; assumptions, 278; capture of plant diversity, 128; completeness of, 8, 55, 152; Daubenmire's impact on, 24; distribution of points, 256; double, 80; error, 277; improving efficiency, 86; intensity, 307; landscape level, 30; patterns, 36–37, 65–69; random, 37–38; for species inventories, 309–310; time, 126–127, 310, 312; plot design and, 320 for transect methods, 130–131for Whittaker plots, 132, 136; unequal, 151; weaknesses of, 145 sampling design, 46, 62, 284; choice of, 125; commonly asked questions, 47; considerations in, 274–275; evaluating with species accumulation curves, 320–321; number of sites, 89; selecting, 111–112; site selection, 33–34, 36, 115, 147; testing in diverse landscapes, 142 sampling techniques, 208; acceptance of, 3; analysis of, 284; calibrating, 317; characteristics of, 129; cost efficiency, 134–137; effect on dominance cover measurements, 131–133; evaluating, 122–125; evolution of, 136; in exclosure studies, 193–194; for foliar cover, 213; inertia to change, 286; of Mueller-Dombois and Ellenberg, 37; random application of, 115; of Raunkaier, 20; shifting, 117; testing alternative, 114 sand; effect on species richness, 206; invasibility and, 234 satellite data, 143–144; use to select sampling sites, 147; vegetation maps based on, 146 scale, xii, 48, 49–50, 269; correlation with species richness, 98; effect on biodiversity,




145; effect on results, 110; of invasibility studies, 174; landscape to regional variability, 11–12; plant to landscape variability, 10–11; plant to plant neighborhood variability, 9 scatter diagrams, 240 scatterplots, 227 Schouw, J. F., 17 Schröter, C., 36–37 searching; areas, 8; around transects, 130; techniques, 73–74; time, 93 seed banks, 186, 234 seed dispersal, 186 seedling establishment, 106 semilog relationships, 101; for describing species-area relationships, 99; between number of species and quadrat area, 100; plot design and, 114 Shannon's index, 199 Shannon-Wiener index, 78 shrub frequency; effect of grazing on, 203–204, 212–213; in exclosure studies, 207 shrublands, exotic species in, 225 Sigmatist school, 21 Simpson's index, 199 single-scale plots, 74–77, 308; to assess plant diversity, 78–79; benefits of, 88–90; limitations of, 90–91, 141 site condition, 26 site-specific variability, 126–127 slopes, of regression lines, 240 Smithsonian Institution, 106; Man and the Biosphere Biological Diversity Program, 277–278 sociability, 21, 27–28 societies, 22 soil analysis, 176 soils; cover, 126; cryptobiotic crusts, 218–219, 225–228, 235, 244; of different vegetation types, 180–181; effect of grazing on, 199–200, 204, 211; effect on species richness, 177, 182, 205–206, 225–228, 243; fertility and species richness, 214–216, 227, 241, 300; mechanical damage to, 216; texture, 206, 208, 225 spatial analysis, 89, 254, 266–270 spatial autocorrelation, 110, 254; assessing from regressions, 261; best models to use with, 264; correcting, 68; effect on cost efficiency, 121; effect on species-area curves, 65; low variance and, 115; subplot overlap and, 114–115; in transect methods, 128, 130; in Whittaker plots, 101 spatial correlation, 260, 261, 267–268 spatial integration, 261–262 spatial interpolation, 264 spatial models, 259; to predict distributions, 321; validating, 266 spatial patterns, use to detect noxious weeds, 168, 170 spatial relationships, 255

spatial scale; in multiscale sampling, 112; in Whittaker plots, 100 spatial variation, 11, 265; in exclosure studies, 212–214; grazing and, 215; means of assessing, 195; scope of, 10; temporal variation and, 8 speciation, lag times, 12

Index species; complementary, 9; as primary units of evolution, 42; renaming, 5 species abundance; Clement's work on, 18; habitat conditions and, 68; index of, 199; negative exponential distribution, 44; plot design and, 52, 115; quadrat area and, 100; underestimating, 134; vegetation type and, 153 species abundance curves, shape of, 7 species accumulation curves, 311; to assess taxonomic completeness, 64–65; effect of plot size on, 52–53; to evaluate sampling designs, 320–321; inclusion in monitoring studies, 277; with various plot designs, 134, 314 species additions, 277 species affinity; of understory to overstory species, 301, 303; for vegetation types, 293–294 species-area curves; to assess taxonomic completeness, 64–65; detection of hot spots and, 319; determining appropriate model, 148; generation of, 101, 311–312; inclusion in monitoring studies, 277; with independent samples, 108–109; variation by vegetation type, 149; from various plot designs, 95, 100, 134, 315–316, 317 species-area relationships, 44, 92–93; accuracy, 316; assessing, 103, 109; scale and, 149; semilog relationships and, 99; use of nested quadrats to measure, 93; vegetation type and, 115 species composition, 48; evaluation in exclosure studies, 198–199; quantifying, 18 species composition overlap, 11, 30, 83, 148, 156; among vegetation types, 151–152; analysis of, 103, 109; in exclosure studies, 198, 204–205; in grasslands, 131; habitat heterogeneity and, 85; role of ecotones, 303; spatial scale and, 212; between vegetation types, 293; within vegetation types, 150–151 species density, 209 species detection. See detection species-environment relationships, 15–16; monitoring, 281 species frequencies, 26; in ecotone studies, 297 species frequency curves, shape of, 7 species inventories; contribution of various plots designs, 314; developing, 73; difficulties of, 307 species relationships, 25 species richness, 69. See also exotic species richness; native species richness; across state plots, 163; advantage of using multiple scales, 103; after fire, 280; analysis of, 115; in arid environments, 223–225; comparable data on, 3; cover and, 234; effect of autocorrelation on,




130; effect of quadrat size, 20; effect of soil characteristics, 205–206, 214–215, 225–228, 235, 240–241; effect of vegetation types, 152–154; environmental factors, 42, 155, 243; estimates of, 285; in floodplains, 81; of grazed versus ungrazed sites, 212–214; in homogenous and ecotone plots, 293; interpolation of patterns, 258; kriging-generated map of, 264–266; mean, 57; patterns of, 42, 298; peaks, 208; plot design and, 114, 316, 319–321; relation to foliar cover, 166; sampling and, 43, 127–128; scaling and, 98; single-scale plot study of, 78–79; in space and time, 44; underestimation of, 136 species turnover, 234; role in invasibility, 188, 190 stability, 22, 35, 235; community, 188 stand, definition of, 25 standard errors, 266 standardized partial regression coefficients, 243 starting locations; effect on species-area curves, 95; influence of, 112; for randomization of plots, 106 statistical analysis, 176, 221–222; selecting appropriate, 252; software, 249, 252; algorithms of, 256; of tree species distributions, 291–292 statistical power, 62 statistics; used by Oosting, 23; use in plant ecology, 20 Stebler, F. G., 36–37 stepwise regression, 232, 242, 259, 270, 292 stratification, causes for difficulty in, 25–26 stratified random sampling, 30, 36, 37, 67, 82, 276 stress tolerators, 18 structural equation modeling, 243, 302 subdominant species, capturing, 300 subjective sampling, 66 subplots; clustering, 112; independence of, 116; sizes, 113; size-shape interactions, 114 subspecies, 4 substrate age, species richness and, 43 succession, 20, 31, 35; assumptions about, 79; chronosequence studies of, 275–276; Clements ideas on, 18; theory, 188; Warming's view of, 17 sun angle, 43 surveys; early, 17; expense and timing of, 5–6; fineresolution, 157; of invasive species, 171; partially eaten plants, 7; perspective on, 11; understanding patterns in, 267 synecology, 34 systematic sampling, 32, 66 Tansley, A. G., 22 taxonomic completeness; statistical techniques to determine, 56; using species accumulation curves to assess, 64–65

Index taxonomic keys, 5 taxonomic resolution, 4 taxonomy, 21; problems in, 4–5 temperature; effect on dominant species, 12; effect on invasive species, 187; effect on species richness, 182–183, 206, 215, 225, 243, 301–302; plant distributions and, 17 temporal plant diversity studies, 277–278 temporal variation, 11, 131; Daubenmire's observations, 31; spatial variation and, 8 Thematic Mapper satellite data, 143 Theophrastus, 16 timber resources, monitoring, 159–160 time series analysis, 274 training, taxonomy, 5 transects, 76–77; belt, 22, 76; Daubenmire, 118–120, 122–125, 128; long-term, 283; methodsadditional searching, 130 capture of plant diversity, 128 cost efficiency, 134–137 detecting exotic species, 132; Parker, 118, 120, 122–125, 128 transportation, migration and, 12 travel time, 136, 312; evaluating, 310 tree diversity; cataclysmic changes in, 278; monitoring, 280–283; sampling to assess, 290–291 tree species richness, 83 tree volume, sampling, 30 type I and type II errors, 60, 62 uncertainty, 278 undersampling, 88 understory species, 155; evaluating, 160; relationship to environmental factors, 292, 294–296 uniqueness, relative, 153 universal transverse mercator location, 125 U.S. Agricultural Resource Service, 123 U.S. Army, Land Condition and Trend Analysis Program, 80, 278–279 U.S. Forest Service, USDA; Forest Health Monitoring Program, 87, 106–108, 159–160, 278–279; modification of Daubenmire transects, 119 U.S. Natural Resource Conservation Service, PLANTS database, 162 variables; biological relation of, 231; identifying, 249–250; independent, 249, 259 variance; emphasis on reducing, 120; spatial autocorrelation and, 115 varieties, 4 vegetation composition, retrospective studies of, 274 vegetation gradients, 67 vegetation maps, 39–40, 146 vegetation patterns, 246; of change, 31; effect of minimum mapping units on, 152; modeling, 301 vegetation types, 3–4, 145, 178–179; effect on species-area

389 curves, 149; high resolution, 147; invasibility and, 233; mapping and classification, 32–33; peripheral areas, 288